Design Approaches in Industrialized House Building

L ICE N T IAT E T H E S I S

Department of Civil, Environmental and Natural Resources Engineering Division of Industrialized and Sustainable Construction

Luleå University of Technology 2017

Emma Viklund Design Approaches in Industrialized House Building

ISSN 1402-1757 ISBN 978-91-7583-888-5 (print) ISBN 978-91-7583-889-2 (pdf)

Design Approaches in Industrialized House Building - A Creativity Perspective

Emma Viklund

Construction Engineering and Management

LICENTIATE THESIS

DESIGN APPROACHES IN INDUSTRIALIZED HOUSE BUILDING A CREATIVITY PERSPECTIVE

Emma Viklund Luleå, May 2017

Division of Industrialized and Sustainable Construction Department of Civil, Environmental and Natural Resources Engineering Luleå University of Technology SE - 971 87 LULEÅ www.ltu.se/shb

Printed by Luleå University of Technology, Graphic Production 2017 ISSN 1402-1757 ISBN 978-91-7583-888-5 (print) ISBN 978-91-7583-889-2 (pdf) Luleå 2017 www.ltu.se

Preface

My two and a half years as a PhD student have been like entering an amusement park full of roller coasters. Each roller coaster led to a new, more complex one, with higher climbs, deeper falls, and considerably more twists and turns. It has often been difficult to foresee where each ride ends up, and what challenges to go through before arriving there. But I have discovered that the more challenging the roller coaster, the more interesting the ride. This journey would not have been possible without the funding of the research project Attract (Attractive and Sustainable Cities in Cold Climates) by the Swedish Governmental Agency for Innovation (VINNOVA), Hjalmar Lundbohm Research Center, and the participating companies. Thank you for enabling my entry into this interesting new place. In addition, I was given access to one particularly exciting roller coaster by Seth M Kempe Foundation Scholarship Fund at Luleå University of Technology. Thank you for the scholarship that funded my conference and workshop journey to Canada. When I entered the amusement park, I had the great luck of having my two supervisors, Helena Lidelöw and Gustav Jansson, by my side. Helena has been a role model since the first time I met her, almost 8 years ago, and remains one still. Thank you for always giving clear answers, strong support, and feedback that has made me develop my research skills and myself as a person. I am happy to now have my role model as a friend. I am also happy to continue working with you, however in a different context. To Gustav, I have this entire adventure to thank, since you were the one who opened the door and convinced me to take the first step over the threshold. For this I am eternally grateful. You have constantly encouraged me to follow my own path, while challenging me

I

to think one step further. I can now stand tall on my own two feet by the exit sign, thanks to your support and friendship. During my journey through the park, I have met some wonderful people that have made the journey enjoyable. Thank you, my coworkers at the division of Industrialized and Sustainable Construction, for your company and all of our interesting discussions. I would like to write a sentence about each one of you, but will restrict myself in order to not write a novel (which I seem to be doing anyway, as usual). I will miss you all. I would however like to specially thank Brian, Ida, Jani and Jutta (alphabetical order!) for always being there through the ups and downs of the bumpy ride. You are superstars. For some of the roller coaster rides I had other PhD students sitting next to me. Me and Martin went on an interesting ride combining social sustainability and the design of houses. I have really enjoyed our teamwork. Angelica and Lisa took me on an innovative adventure, sparking creativity. It is largely thanks to you that my thesis looks the way it does today. Thank you for the inspiration. I also found a kindred spirit all the way over in Australia. Thank you Duncan for showing me that I was not alone in my path between architecture and industrialized house building. I would also like to thank all the inspiring people from the industry who have contributed with their time and knowledge to this research. This would not have been an adventure without you. Thank you for sharing your insights, sitting through timely interviews and showing me your work. Also thanks for your valuable feedback (and sometimes critiques) that have pushed my research forward. You are the ones who made sure that the wagons did not derail. Last but not least, I would like to thank my family and friends for encouraging me, for listening to my stories, and for making sure that not all of my time was spent within the walls of the park. There are other worlds too in need of exploration, which I believe have highly contributed to how much I have enjoyed this particular adventure. With this I am looking forward to the future, and the new challenges ahead. Luleå, May 5th, 2017 Emma Viklund

II

Abstract

Industrialized house builders strive towards structure and control of their processes, including design. Such structure is seldom sought for in architectural design practice, where individuality and autonomy are considered essential. This is causing a tension in the construction industry as industrialized house builders and architects strive to improve their collaboration. In this licentiate thesis, a first step towards better understanding this tension is taken by exploring different design approaches from a creativity perspective. The analytical framework builds on a theoretical conceptualization of design approach features, including design task, design process, design organization, and knowledge sharing. A creativity perspective is added using heuristic or algorithmic design tasks; divergent or convergent design processes; autonomy or imposed structure from the design organization; and sole designer or teamwork as the basis for knowledge sharing. This framework is used to analyze four empirically explored design approaches, two focusing on platform development and two focusing on project specific design. Three of these are explored using a case study approach, while the fourth is explored through interviews with multiple architects. The architectural design approach’s features seem likely to facilitate creativity: the design task is mainly heuristic; the design process enables divergence; and the architects have autonomy in how to go about the design process. However, they experience a lack of knowledge sharing, which could further facilitate creativity. The studied standardized design development approach has the opposite features: the design task is mainly algorithmic; the design process facilitates convergence; and there is a clear structure with instructions of how each subtask should be executed. Hence, this design approach is not likely to facilitate creativity (which was also not its intention). The structure has

III

however improved the design team’s knowledge sharing, which is likely to facilitate creativity. The two platform development approaches have a mix of features. Both design tasks are more algorithmic than heuristic and creativity was not expected of the design results. The design processes are clearly convergent. While the house platform development team is multi-functional and work in close collaboration, the design automation platform developers’ knowledge sharing is based on communication with the client’s organization, thus not enabling as many perspectives on the designed product. The design automation platform developers use a support structure with process descriptions and methods to ensure quality, motivated by the variety of platforms that they develop on a regular basis. The house product platform developers on the other hand, were expected to develop only this one platform. Hence, they worked autonomously, using the teamwork setting to progress and converge in the process. In summary, the creative intention seems related to how design is approached. If creativity is sought for, the design task is heuristic, and divergence and autonomy is promoted. If creativity is not sought for, structure seem to facilitate other benefits such as reliability and quality control. Knowledge sharing could however be potentially beneficial in most design approaches. By understanding that different design approaches will influence creativity in different ways, a first step has been taken towards understanding why architects and industrialized house builders approach design differently.

IV

Sammanfattning

Industriella byggare använder processtrukturer för att få kontroll över processernas innehåll och progression. Detta inkluderar även projekteringsprocesser och processer för plattformsutveckling. Sådan struktur är sällan något som eftersträvas av arkitekter. Deras designprocesser karaktäriseras oftast av individualitet och autonomitet. Skillnaderna i angreppssätt har visat sig skapa motsättningar när arkitekter och industriella byggare strävar efter att förbättra sina samarbeten. I denna licentiatavhandling utforskas olika angreppssätt för projektering från ett kreativitetsperspektiv, som ett första steg i att bättre förstå dessa motsättningar. Analysen bygger på en teoretisk konceptualisering av angreppssätt för projektering och plattformsutveckling. Angreppssätten kan karaktäriseras och analyseras ur ett kreativitetsperspektiv med hjälp av heuristiska eller algoritmiska designuppgifter; divergenta eller konvergenta designprocesser; autonomitet eller införd struktur från designorganisationen; och ensam designer eller teamwork som bas för kunskapsdelning. Två angreppssätt för projektering och två angreppssätt för plattformsutveckling studerades. Tre av dessa studerades genom att göra fallstudier och den fjärde genom en intervjustudie med fem arkitekter. Arkitekternas angreppssätt för tidig projektering verkar främja kreativitet: designuppgiften är heuristisk; designprocessen uppmuntrar divergens; och arkitekterna upplever inte någon påtvingad struktur gällande hur de ske arbeta. De upplever däremot att kunskapsdelningen i processen är bristfällig. Bättre kunskapsdelning skulle ha kunnat ytterligare främja kreativitet i processen. Den standardiserade projekteringen som studerades i en av fallstudierna verkar däremot inte främja kreativitet i någon större utsträckning, vilket inte heller var syftet med angreppssättet. Designuppgiften är i den projekteringsprocessen till

V

största del algoritmisk; designprocessen fokuserar på konvergens; och det finns en tydlig processtruktur med instruktioner för hur varje deluppgift ska utföras. Strukturen har dock lett till en ökad kunskapsdelning mellan företagets olika avdelningar, vilket troligen främjar kreativitet. De två angreppssätten för plattformsutveckling som studerades i varsin fallstudie karaktäriserades båda av designuppgifter som var aningen mer algoritmiska än heuristiska. Kreativitet var inget som eftersträvades och designprocesserna var tydligt fokuserade på konvergens. Vid utvecklingen av husplattformen arbetade ett multifunktionellt designteam i nära samarbete med varandra, vilket främjade kunskapsdelning från olika perspektiv. Vid utvecklingen av designautomationsplattformarna skedde däremot kunskapsdelning främst genom kommunikation med kundens organisation, vilket inte möjliggjorde för lika många perspektiv på slutprodukten. En stödstruktur med beskrivning av plattformsutvecklingsprocessen samt olika metoder för att säkerställa produkternas kvalité används vid utvecklingen av designautomationsplattformarna. Denna stödstruktur motiveras av att plattformsutveckling sker kontinuerligt inom företaget. Vid utvecklingen av husplattformen fanns ingen sådan struktur. Där var gruppen endast tänkt att utveckla en plattform denna enda gång. Baserat på de fyra studerade angreppssätten verkar det finnas en koppling mellan kreativ målsättning och hur man angriper designuppgiften. När man strävar efter kreativa lösningar är designuppgiften heuristisk, designprocessen divergent, och de som arbetar i designprocessen ges autonomitet över hur de vill arbeta. När man inte strävar efter kreativitet verkar struktur användas för att ge fördelar som beräkneliga processer och kontroll över kvalitet, till viss del på bekostnad av den kreativa potentialen. Att främja kunskapsdelning verkar dock ge fördelar oavsett den kreativa intentionen. Genom att öka förståelsen för hur olika angreppsätt för projektering och plattformsutveckling påverkar förutsättningarna för kreativitet har även förståelsen ökats för de motsättningar som visat sig mellan arkitekters angreppssätt och angreppssätten som används inom industriellt byggande.

VI

List of Appended Papers

Paper I

Jansson, G. & Viklund. E. (2015) Advancement of platform development in industrialised building. Proceedia Economics and Finance, Vol. 21, pp. 461-468

Paper II

Jansson, G., Viklund, E. & Lidelöw, H. (2016) Design Management using Knowledge Innovation and Visual Planning. Automation in Construction, Vol. 72, No. 3, pp. 330-337

Paper III

Viklund, E., Sandberg, M., Lidelöw, H. & Jansson, G. (2016) Modularization Based on Commonalities in House-Building Requirements. ICCREM 2016: International Conference on Construction and Real Estate Management, 30 sep-1 oct, 2016, Edmonton, Canada

Paper IV

Viklund, E. (2017) Arkitekters syn på kravhantering vid byggnadsutformning – Resultatredovisning. Technical Report, ISSN 1402-1536, Luleå University of Technology

VII

VIII

Contents

1

INTRODUCTION ..................................................................................... 1 1.1 Background ...................................................................................... 1 1.1.1 Creativity in house building design ...................................... 2 1.1.2 The design process in industrialized house building ............ 3 1.2 Purpose and research scope .............................................................. 4 1.3 Delimitations .................................................................................... 5 1.4 Appended papers and author contribution ....................................... 6

2

METHODOLOGY .................................................................................... 7 2.1 Research design ................................................................................ 7 2.2 Selected design approaches .............................................................. 8 2.3 Empirical explorations ................................................................... 10 2.3.1 Case studies ........................................................................ 11 2.3.2 Interviews ........................................................................... 11 2.3.3 Methods for collecting the empirical material ................... 11 2.4 Analysis .......................................................................................... 14 2.5 Assessing research quality ............................................................. 15 2.5.1 Generalizability .................................................................. 16 2.5.2 Researcher’s background ................................................... 17

3

THEORETICAL FRAMEWORK ........................................................... 19 3.1 Creativity ........................................................................................ 19 3.2 Design approach features ............................................................... 21 3.2.1 Design task – heuristic or algorithmic tasks ....................... 24 3.2.2 Design process – divergence or convergence..................... 26 3.2.3 Design organization – autonomy or imposed structure ...... 28 3.2.4 Knowledge sharing – sole designer or teamwork .............. 30

IX

4

SUMMARY OF RESULTS .................................................................... 33 4.1 House product platform ................................................................. 33 4.2 Standardized design development.................................................. 34 4.3 Design automation platforms ......................................................... 35 4.4 Architectural design ....................................................................... 36

5

ANALYSIS ............................................................................................. 39 5.1 Design approach to develop a house product platform .................. 40 5.2 Standardized design development approach .................................. 41 5.3 Design approach to develop design automation platforms ............ 43 5.4 Architectural design approach ....................................................... 45

6

DISCUSSION AND CONCLUDING REMARKS ................................ 47

7

FUTURE RESEARCH............................................................................ 51

BIBLIOGRAPHY ............................................................................................. 53

X

Introduction

1 INTRODUCTION

1.1

Background

Industrialized house building has gained increased research and industry attention the last 20 years (Lessing et al., 2015) based on the need to improve the construction industry and facilitate house production (Apleberger et al., 2007). Many industrialized house builders use platforms, i.e. “the collection of assets that are shared by a set of products” (Robertson & Ulrich, 1998, p.20), as the foundation to standardize work and increase productivity and quality (Bonev et al., 2015; Jansson et al., 2014). A platform in industrialized housebuilding can include both product and process definitions (Jansson et al., 2014), logistics (Bonev et al., 2015) as well as knowledge, people and relationships (Robertson & Ulrich, 1998). It is however important to balance commonality with distinctiveness in order to enable required customer influence on the final product (Bonev et al., 2015; Jansson et al., 2014; Lidelöw & Olofsson, 2016; Winch, 2003). This balance can be described using different levels of predefinitions within an engineer-to-order (ETO) production strategy (Johnsson, 2013; Wikner & Rudberg, 2005; Winch, 2003). The level of predefinition depends on the customer order decoupling point, i.e. how much of the design that is predefined or produced when the customer places an order (Lidelöw & Olofsson, 2016; Wikner & Rudberg, 2005; Winch, 2003). Regardless of the level of predefinition, Johnsson (2013) found in her multiple case study of Swedish industrialized house builders, that there are benefits from using platforms. These benefits can include e.g. output increase and cost benefits (Johnsson, 2013) as well as increased quality and better control over deliveries (Lessing et al., 2015). In order to reach these benefits, industrialized house builders strive towards structure and control of their processes, including

1

Design approaches in industrialized house building

design (Jansson et al., 2014). Such structure is seldom sought for in architectural design practice, where individuality and autonomy is considered essential (Lawson, 2005). The lack of structure has shown to risk resulting in end-products that are not consistent with their initial intention (Kiviniemi, 2005). In order to decrease this risk, facilitate collaboration and enable clear delivery agreements, some industrialized house builders suggest that structure would also be beneficial in architectural design. 1.1.1 Creativity in house building design However, a reflection should be made regarding how different design approaches influence designers’ creativity. While industrialized house builders use product predefinitions to increase reliability of projects in terms of speed, cost and quality (Johnsson, 2013; Lessing, et al., 2015), architects manage novelty and project distinctiveness in their design processes (Lawson, 2005). Given the standard definition of creativity, i.e. that creativity is the generation of something that is judged to be novel and useful in the intended field (Sawyer, 2012), it is reasonable to expect that the creative intentions in these two contexts are different. A creativity perspective is also relevant from an organizational perspective since creativity has shown important when it comes to employees’ performance, productivity and job satisfaction (Forgeard & Kaufman, 2016). Though some of the benefits from creativity can be seen on an organizational level, creativity originates from individuals (Forgeard & Kaufman, 2016; Hirst et al., 2009). It might seem that creativity is something inherent in certain people, like a personality trait or an inherent talent, but many researchers have concluded that creativity can be facilitated by a person’s context (Hunter et al., 2007). Contextual factors have been shown to affect creativity by influencing a person’s motivation towards the task (Amabile, 1996). When it comes to designing, factors such as task characteristics, structure of the design process and collaboration between design participants can facilitate or restrict creativity (Le Masson et al., 2011). Le Masson et al. use a conceptualization of distinguishing features of design approaches in their discussions of creativity issues. The term design approach is in this thesis defined as “forms of organizing collective design activities”, based on Le Masson et al. (p.218). A design approach can be distinguished by the features design task, design process, design organization, and knowledge sharing (Le Masson et al.), see section 3.2.

2

Introduction

1.1.2

The design process in industrialized house building

The outset for a house design process is a building program where customer intentions are specified (Stockholms byggmästareförening, 2014; Kiviniemi, 2005). The design process then starts with conceptual design and continues through design development (including systems design, and detailed design, see Figure 1.1) (Stockholms byggmästareförening, 2014; Kiviniemi, 2005). Multiple stakeholders are involved throughout the process, iterating and integrating their design results to arrive at the final solution (Hansson et al. 2015). Each project has its own project organization, depending on project type and size, as well as contractual form (Håkansson & Ingemansson, 2013; Lidelöw et al., 2015). Project participants are therefore seldom the same in multiple projects (Lidelöw et al., 2015). Due to the complexity of building projects, the Swedish building industry has become fragmented with narrow specializations (Lessing, 2006; Lidelöw et al., 2015). In addition, coordination is often lacking, resulting in sub-optimization by stakeholders (Apleberger et al. 2007; Lidelöw et al., 2015). Architects are mainly responsible for the aesthetical and spatial qualities of buildings, while technical and production aspects are managed by the contractor’s engineers and/or construction consultancy firms (Lidelöw et al., 2015). Whether or not the architect is able to continue monitoring the design after the conceptual design stage depends on the property developer’s procurement process (ibid.). As follows, there is a clear distinction between conceptual design where architectural perspectives are developed, and design development where engineering perspectives are developed. This is visualized in Figure 1.1.

Figure 1.1. How architectural and engineering perspectives are developed in relation to a general house design process. By predefining building systems and technical solutions in product platforms, parts of the design process can be performed as platform development (Jansson et al., 2014; Lessing, 2006). These platforms often contain system specifications, but they can also contain varying degrees of architectural design and technical details (Viklund, 2015). The product platform part visualized in

3


Figure 1.2 can therefore have a varying span. Figure 1.2 shows the separation of the development of product platforms (blue) and the design process in individual projects (green), based on Jansson et al. (2014). Though the product platform is developed separately, it is applied in projects and integrated as a design constraint in building projects (Jansson et al., 2014; Viklund, 2015).

Figure 1.2. The use of product platforms in a general house design process. Even though the design process can be described in general terms according to Figure 1.1 and Figure 1.2, there are many ways in which designers can approach a design task. Two perspectives on designing for industrialized house building can be identified in these figures: 1) designing the content of a product platform, and 2) designing buildings in individual projects based on platform predefinitions. This thesis presents two design approaches from each of these two perspectives. By including both perspectives, situations where design solutions are meant to be repeatable, as well as situations where design solutions are meant to be unique can be explored. 1.2

Purpose and research scope

The purpose of the research presented in this licentiate thesis is to increase the understanding of how design approaches in industrialized house building influence creativity. To attain this purpose, the features of four different design approaches used in industrialized house building are elucidated through the lens presented in the theoretical framework. This lens builds on the theoretical conceptualization of design approach features by Le Masson et al. (2011), including design task, design process, design organization and knowledge sharing. The effects of different design approach features on creativity is discussed using creativity research as the foundation. The following two research questions guided the synthesis of this licentiate thesis: RQ1: What are the features of design approaches used in industrialized house building?

4

Introduction

RQ2: How do different design approaches influence creativity in design? This licentiate thesis contributes to the body of knowledge on industrialized house-building by exploring a creativity perspective on design approaches: a perspective that is, to my knowledge, missing in industrialized house building research today. Such perspective provides a foundation for discussing and further analyzing the differences between design approaches, as a first step towards better understanding the tension between architectural design practice and design practice in industrialized house building. In order to gain breadth to the reflection on design approach features and their implications on creativity, four design approaches were included in the study, two focusing on platform development and two focusing on project-based design processes. 1.3

Delimitations

This research is not intended to measure levels of creativity, but to reflect upon design approaches and how they are likely to influence designers’ creativity. Hence, it calls on research findings from other researchers in order to allow such reflection. The influence on creativity is focused on designers’ creativity within a design project, and not on e.g. designers’ personal development as creative individuals. Something important to point out in relation to the presented research is that the level of creativity of a product is not the same as the value of the product. A product can be highly appropriate and useful without being original, hence being valuable without being creative. This research is focused on creativity, and not on the perceived value of the end product. The research intends to show some examples from the Swedish industrialized house building industry. These examples were selected to elucidate a variety of design approaches, spanning from conceptual design to detailed design, and including platform development as well as project-based design processes. The extent of the research is however not intended to represent the entire Swedish house building industry, but rather to provide in depth understanding of some of the approaches that are used in an industrialized house building context. Regarding the architectural design perspective, the research is focused on architects’ perceptions on their design approach in projects where platform predefinitions from clients or industrialized house builders are incorporated into the design. These projects were e.g. apartment buildings, preschools, or other house projects of similar scope.

5


1.4

Appended papers and author contribution

Paper I: Jansson, G. & Viklund. E. (2015) Advancement of platform development in industrialised building. Proceedia Economics and Finance, Vol. 21, pp. 461-468 The paper was written in collaboration between Jansson and me. I conducted the case study at the company as well as prepared and presented the empirical material. The final analysis as well as the compilation of the paper was performed by both authors in close collaboration. Paper II: Jansson, G., Viklund, E. & Lidelöw, H. (2016) Design Management using Knowledge Innovation and Visual Planning. Automation in Construction, Vol. 72, pp. 330-337 Jansson was the main author who coordinated and compiled the paper. I contributed by conducting interviews at the case company and writing parts of the narrative of how the case company developed their design management structure. The narrative also includes longitudinal studies conducted by Jansson. I also contributed as a discussion partner when developing the paper, and by reviewing the paper. Lidelöw contributed with knowledge of the case company’s approach and by reviewing the paper. Paper III: Viklund, E., Sandberg, M., Lidelöw, H. & Jansson, G. (2016) Modularization Based on Commonalities in House-Building Requirements. Proceedings of the International Conference on Construction and Real Estate Management (ICCREM): September 30 – October 1, 2016, Edmonton, Canada As the main author I conducted the case study at the company, prepared and analyzed the material and was responsible for the coordination and compilation of the paper. Sandberg contributed with theoretical knowledge of modularization processes, as a discussion partner, and with the compilation of the paper. Lidelöw and Jansson contributed by reviewing the paper. Paper IV: Viklund, E. (2017) Arkitekters syn på kravhantering vid byggnadsutformning – Resultatredovisning. Technical Report, ISSN 14021536, Luleå University of Technology. As the sole author, I performed the interviews, prepared and summarized the material and was responsible for the compilation of the report.

6

Methodology

2 METHODOLOGY

2.1

Research design

An exploratory research approach was chosen since there is not yet sufficient theoretical understanding of design approaches used in industrialized house building and how these influence designers’ creativity. Existing theoretical constructs, building on the design approach conceptualization by Le Masson et al. (2011), were used to support the exploration, thus not striving to develop new concepts but rather to expand the understanding of the theoretical field and the studied real-life events. Figure 2.1 shows the research design and how the design approaches that were selected for exploration relate to a general house design process. The selection of design approaches is further described in section 2.2. The research design was developed through reflections during the empirical explorations. The initial outset was to explore the field and develop the problem definition based on the notion that there is a tension between architectural design practice and design practice in industrialized house building. This is coherent with Langley (1999) who argues that it is complex to study processes in their real-life contexts, thus making it difficult to in advance specify a clear unit of analysis. Hence, the four appended papers provide four explorations with different theoretical perspectives presented in each paper. A qualitative research approach was adopted throughout the four studies, in order to allow exploration and reflectivity (Yin, 2011). Qualitative research has the benefit of enabling exploration of how meaning is ascribed to real-life situations, including the often intricate contextual setting (Flick, 2007; Yin, 2011). The overarching goal was to increase the understanding of how design processes are carried out and why they are carried out in this way. The notably different features of the

7


studied design approaches raised the question of what implications they might have on designers’ creativity. This reflection led to the formulation of the research purpose and research questions presented in this thesis. A second analysis of the empirical material was then made, based on the theoretical framework of design approaches and creativity presented in this thesis, see Figure 2.1.

Figure 2.1: Research design. 2.2

Selected design approaches

Four design approaches were selected to include both the perspective of platform development and the perspective of project-based design. The four empirical explorations span the entire house design process, from conceptual design to detailed design, Figure 2.1. The design approach used to develop the house product platform presented in Paper I spanned from ideation to completion of construction documents, red in Figure 2.2. This case was selected due to the platform’s high degree of preengineering, which includes also architectural perspectives. The high degree of pre-engineering enables the case company to sell the house as a configurable product with pre-defined apartment layouts, aesthetical expression, building system, and technical details.

8

Methodology

Figure 2.2. Scope of the design process when developing the house product platform presented in Paper I. In Paper II, a standardized design development approach at an industrialized house builder was explored. In their design development process, they combine architectural designs from external architects with the constraints from their product platform. Their platform contains specifications of the building system that is used in their factory production. The scope of their design process is visualized in red in Figure 2.3. This design approach was selected due to the company’s clear strive towards standardization of their processes, thus providing a suitable example of a design approach in an industrialized house building context.

Figure 2.3. Scope of the design process of the standardized design development approach presented in Paper II. The design automation platform presented in paper III includes definitions of building elements that are used to automate the production of construction drawings. The design engineers use previously tested solutions to define the platform products. Hence, systems design is only to some extent included in the process, Figure 2.4. Their main work is focused on detailed design, synthesizing previous solutions into platform solutions that will work for design automation. The company has a clear ambition to work consciously with requirement management in their design processes. They were selected for the research for this reason and due to their focus on design development.

9


Figure 2.4. Scope of the design process when developing the design automation platforms presented in Paper III. In Paper IV, an architectural design approach was explored. The study focused on the conceptual design stage and on projects where industrialized house builders’ product platforms were incorporated into the design, Figure 2.5. This design approach was selected to provide an architectural perspective on design approaches.

Figure 2.5. Scope of the architectural design process presented in Paper IV. 2.3

Empirical explorations

In Papers I-III, case studies were conducted in order to enable the study of design approaches in their original settings, not excluding any contextual factors that might be of relevance. All three cases focus on collective design activities, including more than just one designer. To contrast the engineering design perspectives in Papers I-III, Paper IV focuses on architectural design in the conceptual design stage. Since architectural design approaches are often individual (Lawson, 2005), five architects were interviewed to gain multiple perspectives on the approach. Though the architectural design approach is not collective in the same sense as the design approaches in the case studies, they do rely on other stakeholders’ influence for making decisions about the design task and its solution.

10

Methodology

2.3.1

Case studies

Case studies can be used to generate holistic understanding of particular cases, focusing on phenomena within the defined boundaries of the case (Merriam, 2009). Various methods can be used to gather empirical information in case studies, e.g. interviews, observations, participatory observations, document studies etc. (Gillham, 2000). The three case studies presented in this licentiate thesis were conducted using interviews as the main source for information. Additional sources, such as documents, observations and demonstrations were used to triangulate the interview results (see e.g. Flick, 2007). 2.3.2

Interviews

Doing interviews is a suitable research method in qualitative research when you want to investigate how your respondents perceive and assign meaning to certain events (Yin, 2011). Hence, interviews were chosen as the main data collection method to explore the design approaches. The interviews were semistructured, having some predefined questions and topics but still allowing the conversation to take different forms depending on the respondents’ answers. In this way, exploration was enabled without in advance defining a narrow unit of analysis that would limit the understanding of the design approaches and their contextual settings. Some important considerations regarding interviews as research method are the potential to direct the interviews in a way that will affect the research results, and the potential of misinterpretations from both the interviewer and the respondents (Yin, 2011). The directional risk was managed by in advance defining non-directional main questions. In this way, the respondents were allowed to, in their own words, reflect upon the topic without being directed towards specific details. Sufficient details were instead ensured by using probes and follow up questions (see e.g. Gillham, 2005/2008). In this way, the risk for misinterpretations could be reduced since the respondents’ descriptions could be clarified and extended. Additional measures to ensure valid interpretations will be further described in the following sections. 2.3.3

Methods for collecting the empirical material

House product platform, Paper I The development of the house product platform had already been completed before my research project started. Thus, in order to understand the process, I

11


had to rely on the memory of the design process participants. In-depth interviews were therefore selected as a suitable method to capture their experiences of the process (see e.g. Gillham, 2005/2008). I was also given access to archival data and to the platform documentation (the case study was conducted in 2014 when the platform was already in full use). This documentation, in combination with some observations at the case company, was used to gain an understanding of the content of the platform and how the platform was used. In this way, interview questions could be defined to make sure that all aspects of the platform were captured during the interviews. The interviews were semi-structured, with main questions prepared in advance but still allowing flexibility in the dialogue and follow-up questions when needed. The respondents were the 2014 project leader, the project owner, a site manager, an installation coordinator, and a building engineer from the consulting architectural office. Two of the platform development participants were not interviewed. They were a site manager (due to a combination of time constraints and the fact that the site manager perspective was provided by one of the other respondents) and the architect (since he had retired before the research study started). Documentation of the actual development process was not available. Hence, triangulation of the results relied on multiple respondents. Respondent control (Merriam, 2009; Yin, 2011) of the transcribed interviews was used to enable the participants to correct any potential miss-interpretations. The responsible managers were also given the opportunity to read and comment the final summary of the results and analysis. The first step of the analysis was to sort the interview material chronologically and in this way identify a sequence of activities. The interview material was then summarized to increase readability of the results. An analysis was made against product development theories to describe the development process from a value perspective. This analysis is presented in Paper I. Standardized design development, Paper II To explore the standardized design development approach, interviews were conducted with two project managers and the Lean coordinator about how the design process structure was defined, developed and implemented. This empirical information was combined with interview material, observations, documented data such as time reports and external consultant invoices, and participatory experiences collected by my co-authors. The interviews were semi-structured, striving to increase our understanding of how the process structure was developed and what implications it has on the

12

Methodology

company’s project-based design processes. By sorting the interview material in chronological order and then summarizing the content, the process of developing, implementing and gradually refining the design process structure could be described. Triangulation was achieved by comparing the descriptions provided by the three respondents, as well as the empirical data collected by my co-authors. The risk of misinterpretation of interview answers was reduced by having three researchers combine their understandings of the studied case, and in this way elucidate any potential uncertainties. Analysis was then made against Lean product development theories to show how Lean strategies can be adapted to fit an industrialized house building context. This analysis is presented in Paper II. Design automation platforms, Paper III The case company presented in Paper III work continuously with developing product platforms for design automation based on their customers’ already existing products. A possible research approach could therefore have been to participate in an actual platform development process and in this way gain first hand access to the empirical material. This was however not possible due to time constraints. Instead, the design approach was explored through in-depth interviews with four of the six platform developers at the company, demonstrations of the work process, observations made during my one week stay at the company, and document study of company templates that are used as support in the development processes. The respondents were selected in collaboration with the responsible managers to include four of the senior staff members with four different perspectives on the design approach: engineering and modularization; programming and CAD; modularization, project management and administration; and modularization, technical management and sales. The two managers were interviewed together due to time constraints on their behalf. Triangulation was pursued through both multiple respondents and through multiple sources of information (Yin, 2011). Semi-structured interviews were conducted, using probes and follow up questions to gains sufficient depth and detail to understand the design approach. The design process description derived from the interview material was then compared to the company templates and to the demonstrations provided by the modularization engineer. Thus, a coherent description of the design approach could be derived. The analysis and description of the design approach was in Paper III supported by theories on modularization processes in order to add to the existing theoretical understanding.

13


In addition to the triangulation made during the analysis phase, one of the managers was consulted to review the final presentation of the study (the draft paper). This review resulted in some clarifications of the result and analysis description and some reflections regarding my critical view on some of the design approach characteristics. Architectural design, Paper IV For Paper IV, instead of performing a case study where the design approach for one particular house design process could be studied, multiple architects were interviewed regarding how they approach a design task in general. The intention of the study was to provide an initial exploration of how architects manage requirements and platforms in their design process. In total, five architects were interviewed at four different architectural offices. At one of the offices, two architects were interviewed together. Though it is important to reflect on how this interview condition might influence the interview answers, it was my experience that the two architects questioned and encouraged each other in a way that challenged their own reflections, thus leading to lively discussions and depth to the empirical data. All interviews were semistructured and the interview protocol contained questions and topics related to how requirements and constraints from clients and product platforms are managed in the design process, and how the architects reflect on topics such as design automation, configuration and parameterization. This study focused on individual architects’ experiences of their own design approaches. Triangulation in the sense of multiple sources was therefore not considered relevant in this study. Yin (2011) argue though, that triangulation can still be achieved by interviewing the respondents on multiple occasions and in this way ensuring that the respondent’s view is correctly represented. Due to the intended use of the study as an initial exploration, this triangulation method was not implemented due the added resources it would require. Instead, research quality was managed by trying to decrease the risk for misinterpretations by allowing thick descriptions during the interviews, and by using probes and follow up questions. The interview material was compiled by sorting the material in categories derived from the material itself. Each category was then summarized. The study is published as a technical report containing the summarized results of the interviews. 2.4

Analysis

In addition to the analyses presented in Paper I-III, an overarching analysis regarding design approach features and their influence on creativity was made

14

Methodology

in this thesis. Based on my increased understanding of the studied real-life events, theory on design approaches and creativity was revisited in order to develop a theoretical framework to support the discussions. The design approach features as conceptualized by Le Masson et al. (2011) provided a good fit with what could be seen in the empirical material and was therefore selected as the foundation for the framework. Research on the differences in engineering design theory and artistic design theory was added to support the discussion on different features of design approaches (e.g. Cross & Roozenburg, 1992; Eder, 2013; Lawson, 2005). Creativity research was then consulted to enable a discussion of how design approach features are likely to influence creativity (e.g. Amabile, 1996; Caniëls et al., 2014; Csikszentmihalyi, 1996). The model of analysis, which is presented in Table 1, can be regarded as a proposition for a framework to reflect on differences in design approaches in industrialized house building and thus support reflections on why different approaches are used by different designers in the industry. In this licentiate thesis, the theoretical framework was used to re-analyze the empirical material from the four appended papers and reflect on the design approaches’ differences and creative implications. 2.5

Assessing research quality

A central part in assessing research quality is to assess the validity (what Flick (2007) calls credibility) of the results (Yin, 2011). Yin (2011, p.78) defines a valid study as “one that has properly collected and interpreted its data, so that the conclusions accurately reflect and represent the real world (or laboratory) that was studied”. To strengthen the validity of qualitative research, transparency in how the research was conducted is one important aspect (Yin, 2011). This was sought by striving towards clear descriptions of how empirical material was gathered and analyzed, as well as reflections on how I as a researcher might have influenced the results and interpretations. Reflections on researchers’ subjectivity and its influence on interpretations are discussed by Gibbs (2007) under the term reflexivity. If not fully included in the presented papers, these reflections were made elsewhere as part of processing the material. A description of my knowledge background can be found in section 2.5.2. By submitting the papers for peer review, other researchers had the opportunity to question the content of the paper and thus provide input that could further clarify the descriptions therein (Flick, 2007). One difficulty that I experienced during the research was to present the research in a clear way in written English, given that this is not my native language. Feedback from peer reviews, as well as respondent feedback, was therefore used to improve and

15


validate the written presentations. Regarding the collection of empirical material, triangulation through multiple methods and multiple respondents was the main approach to strengthen validity. To further validate the results, triangulation could have been used to a greater extent, for example by participating in a development project for Paper III or by gathering more of the documentation that was most likely made during the design process studied in Paper I and when developing the standardized design process in Paper II. However, these efforts would have been extensive, and were not prioritized at this initial research stage. The measures that were used to achieve validity were assessed as sufficient to be able to present an initial interpretation of design approach features used in industrialized house building and how these are likely to influence creativity. Validity regarding how the studied design approaches are likely to influence creativity is however limited to the validity of the theoretical constructs used for analysis, since no measures of creativity have been attempted (see Mumford, 2003). In order to strengthen the validity of these theoretical constructs, creativity influencers were discussed using multiple researches’ findings. Another important consideration is the comparability of the design approaches discussed in this thesis. First of all, platform development and project-based design provides two different perspectives, not directly comparable to each other. They were however selected just for this reason, to include both of these perspectives in the reflections. The two design approaches that in this thesis represent project-based design are also considerably different. While architectural design is focused on individual architects’ self-chosen approaches (i.e. an individual level of creativity (Anderson et al., 2014)), the standardized design development approach provides an organizational perspective (i.e. a team level of creativity (Anderson et al., 2014)). This is in itself an interesting reflection, that in architectural practice, individuality is promoted, while industrialized house builders promote an organizational and hence collective approach towards design. These different levels of analysis, the individual and the team level, can therefore, in the context of the tension between architectural design practice and industrialized house building, be regarded as a research strength rather than a comparability limitation. 2.5.1 Generalizability By not controlling the contextual setting, like when doing e.g. experiments, and instead studying phenomena in their real-life contexts, generalizability becomes limited to situations with similar settings (Merriam, 2009). But by providing substantially rich descriptions of each studied case, readers are

16

Methodology

allowed to by themselves determine which part of the research is applicable in their own setting (Merriam, 2009). Hence, rich descriptions of cases were pursued, however restricted by the page number limitation of the publishers. The results of the research presented in this thesis was used to increase the understanding of how design approaches influence creativity in design, by using an analytical model derived from design and creativity research. Hence, the empirical explorations inform existing theoretical constructs, which in turn can be generalizable to other, similar contexts (Yin, 2011). In this way, some extent of generalizability can be achieved even though each studied design approach is highly individual and set within its particular context. The qualitative approach of the research suggests that the final results should lead to working propositions rather than conclusions (Yin, 2011). Subsequent research can then strengthen or contradict these propositions, thus evolving our understanding of the theoretical constructs further. 2.5.2

Researcher’s background

I started my PhD studies in construction management after attaining a master degree in architectural engineering at Luleå University of Technology. My work experience in construction is limited to a few summer internships focused on traffic planning and also on supervising youths in conceptual city planning projects. I do however have a quite large network in the industry, with friends and relatives working as structural engineers, project managers, infrastructure planners, carpenters, HVAC engineers, and architects. This has enabled at least some insight into the everyday work in the industry. As part of my engineering education, I did one semester at Politecnico di Milano in Italy. There I had the opportunity to attend engineering courses from the civil engineering program as well as a design studio course from the architecture program. I consider myself to have an engineering mind-set when it comes to problem solving: I appreciate structure and am good at planning. But I also have an artistic vein, where drawing, painting, sewing, and interior design projects are continuously ongoing in my spare time. My interest in architecture reaches back as far as I can remember and it is my opinion that architecture is a major contribution to how we perceive an environment. A house is not just a house but it contributes to the area where it is situated. Hence, architectural qualities are, according to me, important to consider and prioritize when designing and building houses.

17


Based on this background and my personal interests, I consider myself to have one foot on each side of the interface between an architectural perspective and an engineering perspective on the house building industry. My initial expectation of this research project was to develop methodologies for improved structure in architectural design that could facilitate collaboration with industrialized house builders. This expectation has however evolved as the research progressed, and I now find it more relevant to first gain a rich understanding of the implications of different design approaches. My perception is that structure is positive in certain situations, but it can also entail consequences to a creative process. It is therefore important to increase our understanding of the effects of design approaches in relation to their intentions.

18

Theoretical framework

3 THEORETICAL FRAMEWORK

3.1

Creativity

According to the standard definition of creativity, in order for something to be regarded as creative, the product (i.e. the result of the creative process) should be both original and effective in the intended domain (Batey, 2012; Forgeard & Kaufman, 2016; Howard et al., 2008; Mumford, 2003; Runco & Jaeger, 2012). Originality in this sense means something unusual, novel or unique; and effectiveness means something useful, appropriate or valuable (Runco & Jaeger, 2012). Csikszentmihalyi (1996) proposed a systems model in his research on creativity, claiming that creativity cannot be separated from its recognition within the field in which it is supposed to be creative. A creative piece of art will e.g. not be considered creative unless it is exposed to the art field. Thus, his systems model includes domain, field and person. The domain is the symbolic system (i.e. vocabulary or conceptual organization of rules and procedures) within which a person aims to creatively contribute. The field is the group of people who jointly decide whether the creative contribution will become part of the culture within the domain. In accordance with this view, Sawyer (2012) defines creativity as “the generation of a product that is judged to be novel and also to be appropriate, useful, or valuable by a suitably knowledgeable social group”. Hence, creativity can be described as generating something that is judged by the field as creative in the intended domain. The construct of creativity is complex and multifaceted (Anderson et al., 2014; Caniëls et al., 2014; Sawyer, 2012). It has been researched from various perspectives (Batey, 2012) such as genetics (e.g. Kandler et al., 2016), psychology (e.g. Csikszentmihalyi, 1996), management (e.g. Amabile, 1996; 1997), and neuroscience (Sawyer, 2011). For overviews of the historical

19


development and current research perspectives on creativity, see Anderson et al. (2014), Kaufman & Sternberg (2010), Mumford (2003; 2012), Runco (1997), Sawyer (2012), Shalley et al. (2004) and Sternberg (1999). The variety of research perspectives and approaches has contributed to develop a rich, however somewhat diverse (Batey, 2012), understanding of the creativity construct (Mumford, 2003; Sawyer, 2012). This includes the recognition that creativity is linked to various factors such as personality traits, cognitive style and contextual setting (Liang et al., 2013; Shalley et al., 2004; Shalley et al., 2009). The links between these factors are often complex and studies have shown, for example, that some contextual factors seem to have different influence on different people (Caniëls et al., 2014; Choi et al., 2009). Such intricate relations are therefore important to recognize when discussing creativity. However, when reviewing creativity research, there seem to be a consensus regarding some creativity influencers. Hence, a discussion on how creativity is influenced by contextual factors is enabled, while staying mindful of the complexity of the creativity construct. One consensus model of creativity is that contextual factors affect individuals by influencing their intrinsic motivation (Amabile, 1996; Bergendahl et al., 2015; Hon, 2012). This theory was initially developed by Amabile (1983), and has since been tested and refined by multiple researchers, leading to the “proposition [being] elevated to the status of an undisputed principle” (Hennessey, 2003, p.255). Intrinsic motivation means the kind of motivation that makes a person passionate about a task for the sake of the task itself, and not for gaining extrinsic affirmation (Amabile, 1996). Contextual factors supporting autonomy, competence or task involvement are likely to positively influence intrinsic motivation (ibid.). This positive influence has shown to be higher if a person’s initial level of intrinsic motivation is high (ibid.). Contextual factors that are experienced as controlling are on the other hand likely to reduce creativity (ibid.). Both Amabile (1996) and Caniëls et al. (2014) suggest that different stages of the creative process can be facilitated by different kinds of socialenvironmental factors. During idea generation, intrinsic motivation plays a predominant role (Amabile, 1996). Caniëls et al. (2014) found that extrinsic rewards should not be used during this initial stage, since this might hinder creativity. They also found that creativity during idea generation can be facilitated by a non-hierarchical organizational setting where individuals with different knowledge and expertise can interact in a ‘safe environment’, working on challenging tasks. Sufficient resources to allow interpersonal contacts and access to information was also shown positively related to

20


creativity in the idea generation stage. At the idea implementation stage, synergistic extrinsic motivators, such as e.g. external rewards, can facilitate creativity by making the creative individuals focused on finalizing their ideas (Amabile, 1996; Caniëls et al., 2014). The group should then be composed of individuals with complementary expertise and implementation know-how (Caniëls et al., 2014). The team leader is at this stage crucial. This person should have a hierarchical position and be able to make decisions and bare the final responsibility (ibid.). Resources such as funds, time and competencies were also found by Caniëls et al. to positively influence creativity in the idea implementation stage. Caniëls and her colleagues provide an interesting perspective on the different stages of a creative process, and how different contextual settings might be relevant for idea generation or implementation. 3.2

Design approach features

Design takes place in many different fields of work (Lawson, 2005), and designers in different contexts use different design approaches to arrive at suitable (and sometimes creative) design solutions. Many researchers have distinguished between the theoretical representations of the engineering design process and the architectural or industrial design process (e.g. Cross & Roozenburg, 1992; Eder, 2013; Lawson, 2005; Le Masson et al., 2011). Architectural and industrial design (collectively called artistic design by Eder (2013) and in this thesis) have in common that they include design objectives such as aesthetical expression, user functionality, emotional value, and semantic or symbolic value (see e.g. architectural design: Lawson, 2005; and industrial design: Le Masson et al., 2011). Engineering design on the other hand regards technical systems where the functional behavior can be “fully determined by physical principles and can be described by physical laws” (Cross & Roozenburg, 1992, p.326). The engineering design problem is thus to find solutions that provide the technical system with required properties in the most effective and efficient way (ibid.). Whether engineering design solutions can be regarded as creative has been questioned by some researchers, due to the straight forward (algorithmic) character of the design task (Amabile, 1996). Amabile argues that in order for a design solution to be regarded as creative, the design task has to be open-ended (heuristic) to some degree. Le Masson et al. (2011) do however reflect on creativity issues in engineering design, concluding that some engineering approaches allow certain creativity enablers such as divergent thinking and autonomy while still converging and structuring the process. His description suggests that even though engineering problems can be regarded as more algorithmic than artistic problems, they can still be

21


heuristic to some degree. Lawson (2005, p.4) exemplifies the split between the characteristics of engineering design and artistic design by using fashion design as an example: “The engineer’s process seems to us to be relatively precise, systematic and even mechanical, whereas fashion design seems more imaginative, unpredictable and spontaneous. The engineer knows more or less what is required from the outset. In this case a beam that has the properties of being able to span the required distance and hold up the known loads. The fashion designer’s knowledge of what is required is likely to be much vaguer. The collection should attract attention and sell well and probably enhance the reputation of the design company. However, this information tells us much less about the nature of the end product of the design process than that available of the engineer designing a beam.” Both the nature of a design task and the structure of a design process can be recognized in this description. To support creativity discussions related to design approaches, Le Masson et al. (2011) distinguish between three decisive features: the design process, the design organization, and knowledge sharing1. The design task is not directly specified by Le Masson et al. as one of the features, but their research (as well as the research by e.g. Amabile (1996) and Lawson (2005)) clearly shows that the design task has a relevant influence as well. Building on this conceptualization, four distinctive design approach features can be identified to support the discussion on how creativity is influenced, Figure 3.1. How the design organization influence the design process was also recognized by Jansson et al. (2014) in an industrialized house building context. They proposed collaborative design (i.e. knowledge sharing) as one support method, Figure 3.1. The four design approach features can be recognized in creativity research as potential influencers on creativity. Research has shown that the nature of a design task, if it is heuristic or algorithmic, affects the level of creativity that will be achieved (see e.g. Amabile, 1996; Csikszentmihalyi, 1996; Hunter et al., 2007). The design process can be described as either divergent or convergent (or a mix of the two) where divergence has shown essential for idea generation while convergence has shown essential for idea implementation (see e.g. Caniëls et al., 2014; Howard et al., 2008; Le Masson et al., 2011). 1

Le Masson et al. discuss knowledge acquisition and management as this component, but by using teamwork as the foundation for discussing how knowledge is made available in the design process, the more narrow term knowledge sharing is used in this licentiate thesis. This term is further defined in section 3.2.4. 22


Regarding the design organization, there are numerous factors that are likely to affect creativity, e.g. resource availability, co-worker relationships, or structure etc. (Amabile, 1996). One important factor is how autonomy or imposed structure affects a persons’ intrinsic motivation towards the task, and thus their creativity (Amabile, 1996; Chang et al., 2012; Choi et al. 2009; Oldham & Cummings, 1996). How knowledge is made available in a design process can also influence creativity, since new viewpoints can inspire idea generation (Amabile, 1997; Basadur & Gelade, 2006; Csikszentmihalyi, 1996). Multifunctional teamwork is one example that can facilitate such knowledge sharing (Carlisle & Dean, 1999; Gilson et al., 2005). In Table 1, the design approach features of Figure 3.1 are polarized using artistic design and engineering design as the two sides of the scale. Though such theoretical polarization can be identified in design theory, Carlisle & Dean (1999) argue that they should not be considered as alternatives: design can benefit from having a mix of both approaches. Empirical investigations have also shown that in practice, design approaches are seldom fully towards only one of the two sides (Cross & Roozenburg, 1992). I would also like to point out that not all artistic design is creative nor is all engineering design uncreative. The polarization is however useful to support design approach reflections.

Figure 3.1. Conceptualization of design approach features, based on Le Masson et al. (2011) and Jansson et al. (2014).

23


Table 1. Design approach features from a creativity perspective Design task

Design process

Design organization

Knowledge sharing

ARTISTIC DESIGN Heuristic

Divergent

Autonomy

Sole designer

Algorithmic

Convergent

Structure

Teamwork

ENGINEERING DESIGN

3.2.1 Design task – heuristic or algorithmic tasks There is a clear distinction between how the design task is perceived in engineering design and in artistic design (Cross & Roozenburg, 1992). The engineering design problem is in theory often described using a tree structure, where problems and their subproblems can be specified before synthesizing design solutions (ibid.). Focus of the design task is on physical performance, whereas artistic considerations such as appearance is usually not prioritized (ibid.). Artistic design problems on the other hand, are seen as ill-defined, open-ended problems where problem definition and problem solutions emerge parallel to each other (Cross & Roozenburg, 1992; Eder, 2013; Le Masson et al., 2011). The artistic element means that properties such as appearance, usability, attractiveness, marketing and emotional reactions are important considerations to include (Eder, 2013). Such objectives are however difficult to specify and predict before starting the actual design process since the understanding of a product evolves as the design progresses (Lawson, 2005). Sketching is used to gradually evolve design ideas from vague, tentative conceptions towards detailed representations of design suggestions and embodiments of how the designer perceives and intends to solve the design problem (Javid, 2014). Though Lawson (2005, p.4) distinguishes the differences between an engineering design task and an artistic design task in terms of distinctiveness, he argues that most designers “deal with both precise and vague ideas, call for systematic and chaotic thinking, [and] need both imaginative thought and mechanical calculation”. This might be explained by Howard et al. (2008) who claim that both conceptual design and embodiment design contain analysis, generation and evaluation, i.e. all phases of the creative process.

24


A similar distinction as the one between engineering design and artistic design has been made in creativity research. In order to achieve high levels of creativity, designers should be allowed to reflect on and influence the problem definition (Csikszentmihalyi, 1996; Mumford, 2000; Runco, 2007). Wellspecified problems that are presented to a designer, with little autonomy regarding the content of the task, are likely more easy to solve but will result in outcomes that are less creative (Csikszentmihalyi, 1996). Amabile (1996) reflect on these task characteristics as heuristic (i.e. open-ended) tasks and algorithmic (i.e. closed-ended) tasks. She argues that algorithmic tasks cannot result in creative outcomes, since such tasks do not allow the designer to diverge: the straight forward characteristic promotes one way forward and the problem will be solved accordingly. Based on the notion of intrinsic motivation as the main conductor of creativity (Amabile, 1996; Csikszentmihalyi, 1996; Mumford, 2000), it is advantageous if the designer experience the design task as intellectually challenging (Hunter et al., 2007; Shalley et al., 2004). Such intellectual challenge can stem from both highly undefined tasks, e.g. being given a white sheet to draw on, as well as highly constrained tasks, e.g. welldefined commissioned art-works: as long as the task is still to some degree heuristic (Csikszentmihalyi, 1996). However, too few constraints can result in high complexity leading the designer to reuse ideas in order to make the task more manageable, thus reducing the level of creativity (Caniëls & Rietzschel, 2015). As this discussion suggests, research on task constraints’ effects on creativity has shown mixed results (Caniëls & Rietzschel, 2015): task constraints can be both enabling and restraining (Onarheim & Biskjaer, 2013; Biskjaer et al., 2014). A high degree of constraint does not necessarily result in an algorithmic task. Some constraints might be contradicting and others might be perceived as challenging to solve, thus increasing task complexity, the need for divergence, and the sense of intellectual challenge towards the task. Whether or not a person feels motivated by such challenge is however dependent on that persons’ personality traits and cognitive style (Caniëls et al., 2014; Choi et al., 2009). If the constraints help to clarify project goals, it is likely that positive synergies with intrinsic motivation will be achieved, and thus enhanced creativity (Amabile, 1996). It is also relevant to discuss the level of creativity that is expected from a design task (Howard et al., 2008). If a task is not intended to result in originality for example, it is reasonable to not expect the designer to exhibit creativity, regardless of how well creativity is facilitated by the organizational context (Choi et al., 2009).

25


3.2.2 Design process – divergence or convergence Cross & Roozenburg (1992) describe that there are fundamental differences in the consensus models of the engineering design process and the artistic design process. Engineering design is described as a mainly linear process, however using iterations between process steps (Cross & Roozenburg, 1992; Pahl et al., 2007; Wikberg-Nilsson et al., 2015). Artistic design on the other hand, is described as a cyclic, spiral process where cognitive processes are in focus rather than the sequential progression of the project (Cross & Roozenburg, 1992; Kiviniemi, 2005; Lawson, 2005). An important reflection at this point is the engineering designers’ strive to structure and manage their processes (Le Masson et al., 2011), which can be one explanation to the linearity of the consensus model (Howard et al., 2008). By artistic designers, such control is often regarded as inhibitory rather than positive (Green & Bonollo, 2002; Runco, 2007). Hence, the consensus model of artistic design does not strive to standardize and control the design process, but rather to increase the understanding of what makes design processes more or less successful (Cross & Roozenburg, 1992). This is what Cross & Roozenburg (1992) calls the prescriptive and descriptive characteristics of the respective consensus models. The engineering design approach aims at making complex processes tangible and transparent by structuring the process in steps of analysis and synthesis: each step providing more concrete descriptions of the solutions (Pahl et al., 2007). The process is described as sequential but is not meant to be rigidly interpreted (ibid.). It should be used as a guideline, adapting it to each design situation by removing or reorganizing process steps (ibid.). The output of each step is used as basis in the subsequent step and to analyze design suggestions against the initial (but evolving) requirements (Cross & Roozenburg, 1992; Pahl et al., 2007). According to Howard et al. (2008), some of the engineering design models promote both divergence and convergence, where idea generation is followed by integrated evaluation and selection of ideas (i.e. convergence). However, Le Masson et al. (2011) argue that most of the engineering design theories mainly promote convergent thinking methods. This is in line with Cross & Roozenburg’s (1992, p.327-8) description of the convergent characteristic of the engineering design consensus model: “Firstly, it assumes that design should proceed from the general and abstract to the particular and concrete, i.e. the problem should be analysed in abstract terms, before any material concepts are established, and those concepts are then gradually refined in increasingly more detail. Secondly, it assumes that complex problems should be

26


decomposed into subproblems, for which subsolutions are to be found and 'synthesized' into overall solutions for the design problem […].” In architectural design, both conceptual thoughts on the building’s appearance and thoughts on functional properties need to be developed. These two perspectives are according to Peponis et al., (2015) developed in parallel, using drawings to test the design against conceptual thoughts and develop conceptual thoughts based on the emerging design. Green & Bonollo (2002) suggests a phase-based design process description for industrial design and product development. Lawson (2005) questions this type of sequential representation though, arguing that an artistic design process is built up of iterations between analysis, synthesis and evaluation, developing the problem definition along with the solution suggestions. He claims that heuristic problems call for iterations between the general and the detailed. Details can often be part of the initial idea generation and creative conceptualization. He stresses though that each design process is individual (and shaped by each individual designer), hence making it difficult to theorize how a creative design process will be organized. Individuality in how to go about the design process is emphasized, giving the designer autonomy over how to perform each individual task. In artistic design, divergence is key to explore design possibilities alongside the problem definition (Le Masson et al., 2011). However, convergence is also necessary and integrated as a natural part of the process in order to synthesize feasible ideas into something possible to implement and present (Green & Bonollo, 2002). Regarding divergence and convergence in creativity research, divergence tests have become one standard measure to predict individuals’ creative potential (Runco, 2007). Many researchers have recognized divergence as important in a creative process since it enables exploration of new possible solution paths (Le Masson et al., 2011). Divergence means to consider and generate many different ideas, thinking in multiple directions (Runco, 2007; Tan, 2015). Convergence, on the other hand, is to start from many ideas or possibilities and synthesizing them towards one answer (Runco, 2007; Tan, 2015). Divergent and convergent thinking is linked to cognitive thought processes (Runco, 2007), but some researchers have started to view the creative process as activity-based rather than solely cognitive (Howard et al., 2008). Thus, divergence can be more or less promoted by the structure of a design processes (Howard et al., 2008). Divergence is indisputably important for idea generation, but convergence has also shown important in the creative process in order to synthesize and evaluate ideas (Caniëls et al., 2014; Howard et al.,

27


2008; Le Masson et al., 2011; Tan, 2015). Runco (2007) concludes that heuristic tasks calls for divergent processes, while algorithmic tasks does not. 3.2.3 Design organization – autonomy or imposed structure A design organization can influence designers by e.g. imposing work structures on the design process. Whether or not these kind of structures are facilitators or inhibitors for creativity is somewhat disputed in creativity research (Chang et al., 2012; Bergendahl et al., 2015). Some researchers discuss autonomy as decisive to creativity (e.g. Amabile, 1996 and Oldham & Cummings, 1996) while other researchers have concluded that structure can improve creativity for some individuals (Chang et al., 2012; Choi et al. 2009). Le Masson et al. (2011) distinguish between ‘autonomous creators’ working in creative teams and ‘administrated creativity’ where designers’ autonomy is influenced by the structures imposed by leaders in the organization. Gilson et al. (2005) reflect on a contradiction that can be found in many organizations. On the one hand, organizations strive towards structure in order to manage complexity and ensure consistent operations. But on the other hand, in order for organizations to be creative and cope with uncertainty, risk taking and a will to try new approaches are necessary. Hence, a discussion of how structure and autonomy influence creativity is relevant when trying to understand design approach implications. In engineering design, complex problem solving seems to often be managed by imposing structure (Cross & Roozenburg, 1992; Green & Bonollo, 2002). However, some autonomy in how to perform a design task can still be allowed: “The consensus model does not restrict designers to just one way of working. Instead, it tries to organize the problem-solving behaviour of designers so that this behaviour will be more effective and efficient than intuitive, unaided, unsystematic ways of working.” (Cross & Roozenburg, 1992, p.328) An imposed structure in design can stem from a prescriptive process description or from prescribed methods. Many of the methods that have been developed to support design processes impose a structure to how the design process is to be carried out (Green & Bonollo, 2002). However, some methods are developed to support idea generation while other methods are developed to ensure quality and structure in idea implementation (Green & Bonollo, 2002). Not surprisingly, Green & Bonollo suggests methods focused on divergence

28


for the conceptual design phase and methods focused on evaluation and synthesis (i.e. convergence) for the phases embodiment and detailed design. In artistic design, progressively refined sketches and visual representations are used to develop and communicate design ideas (Eder, 2013; Peponis et al., 2015). Lawson (2005) stresses the individuality of approaches used by designers, since they might have different preferences regarding what methods to use, how to use them and when. In addition, Eder (2013) describes the artistic design process as mainly intuitive, thus often not pre-specifying particular methods to use during idea generation and implementation. This mainly intuitive process is often performed in studio settings, using personal judgment for selecting which ideas to continue developing (Eder, 2013). Cross & Roozenburg, 1992 describe this as an opportunistic process. Accordingly, there seem to be a perception in the more artistic design fields, that imposed methods are likely to impede creativity (Green & Bonollo, 2002; Runco, 2007). Regarding the influence of autonomy on creativity, Hunter et al. (2007) found that autonomy seem to have less effect on design creativity than other organizational factors such as intellectual stimulation and positive interpersonal exchange. There are also moderating factors that influence the effect of autonomy (Chang et al., 2012). Chang et al. found that the effect of task autonomy (defined by them as the lack of pacing and definition of subtasks) on creativity was moderated by prior task experience and self-control. Their findings suggest that autonomy is positive for creativity if the individual has previous experience of similar tasks. Novices on the other hand seem to benefit from clear structure. If the individual is unfamiliar with the task and is given autonomy in how to solve it, creativity was shown to decrease. Chang et al. explain this by suggesting that the feeling of ambiguity might result in individuals feeling uncertain of what to do. Imposed structures are then likely to provide clarification rather than control, hence making the task more manageable and increasing intrinsic motivation. By self-control, Chang et al. mean “the deliberate management of the self, enabling individuals to regulate and alter their inner responses in appropriate ways” (Muraven, Tice, & Baumeister, 1998, in Chang et al., 2012, p.709). They found that people with high self-control seem to perform evenly, regardless of the level of autonomy. People with low self-control on the other hand exhibited less creativity when working on autonomous tasks, while exhibiting increased creativity when being given a structure to follow. Personality traits have also been recognized by other researchers to moderate the influence of autonomy on creativity. Choi et al. (2009) tested the influence

29


of standardization and routinization on creativity. They define standardization as “rules and procedures relating to the task” and routinization as “the nature of the work involved in completing the task” (Choi et al., 2009, p.334). Standardization thus refers to standard operating procedures of how to perform a task, while routinization involves repetitiveness and lack of novelty, likely resulting in a feeling of boredom (Choi et al., 2009). Their research shows that task standardization negatively influence individuals with high creative ability, while individuals with low creative ability showed slightly positive indications. By creative ability, they refer to “skills or competencies relevant to creative performance, such as the ability to generate new ideas or look at problems from novel perspectives” (Choi, 2004, in Choi et al. 2009, p.338). This is in support of Oldham & Cummings (1996) who found that individuals with high creative ability seem to benefit from complex jobs (including autonomy) and non-controlling leadership, while individuals with low creative ability seem to become less creative in such conditions. Task routinization on the other hand, was showed by Choi et al. (2009) to have clear negative influence on creativity, regardless of creative ability. According to Csikszentmihalyi (1996) this negative effect stem from routinization making the task into an everyday chore which is likely to negatively influence a person’s intrinsic motivation. 3.2.4 Knowledge sharing – sole designer or teamwork Knowledge acquisition and integration in the design process has shown crucial when working with complex systems (Carlisle & Dean, 1999; Le Masson et al., 2011). To achieve knowledge sharing between different knowledge domains, Carlisle & Dean (1999) suggests multi-functional team work in a work environment where each designer’s perspective will be equally valued and communicated in such a way that other designers will understand its reasons and implications. In such an environment tacit assumptions are more likely to surface, which can contribute to challenge fixation and generate new ideas. Knowledge sharing is described by Basadur & Gelade (2006, p.47) as “making information widely available”. Teamwork is also promoted by Gilson et al. (2005, p.552, based on Tesluk, Farr & Klein (1997)) who argue that “[w]hen employees work in teams, they are exposed to a broad range of perspectives, skills, and information that they can use to generate ideas and form new and different options about how work should be done”. Multi-functional teams can contribute to bridge the gap between artistic perspectives and engineering perspectives and in this way provide a solid foundation to find compromises between what is desirable from a market perspective and what is technically possible from a structural and production perspective (Carlisle & Dean, 1999). By coordinating architectural, structural and production perspectives, a balance

30


can be reached between functional, aesthetical, economical and technical requirements (Peponis et al., 2015), reducing the risk of suboptimisation discussed by Lidelöw et al. (2015). Such synthesis of different perspectives is potentially the key to successful product design (Carlisle & Dean, 1999), and to knowledge build up and integration of new ideas in the building industry (Håkansson & Ingemansson, 2013). Knowledge of the domain within which an individual is acting is essential for creativity (Amabile 1996; 1997; Csikszentmihalyi 1996). Amabile (1997) discusses expertise as this component, including both knowledge and an inherent talent of the domain. A similar reflection is made by Mumford (2000) who describe that expertise, i.e. knowledge acquired with experience, is foundational to creative problem solving. If you do not know the symbolic system of the domain (i.e. the vocabulary and conceptual organization of rules and procedures), you cannot contribute with ideas or solutions that will be considered appropriate, useful or valuable by the field (Csikszentmihalyi 1996; Mumford 2000). However, some researchers have concluded that knowledge also can create fixations that can reduce creativity (Le Masson et al., 2011). By focusing on existing strengths, one might reduce the chance of finding new paths forward (Carlisle & Dean, 1999). Csikszentmihalyi (1996) describes that the essence of creativity is to connect knowledge together in ways that has not been done before. Thus, knowledge from multiple domains is often beneficial (and crucial) for creativity, since it allows for new connections and viewpoints to be found (Csikszentmihalyi 1996), hence reducing the risk for fixations. This relates to Mumford’s (2000) discussion of how expertise, including knowledge and experience of the domain and of methods for problem solving, may ease the adoption of new knowledge and enable analogical reasoning, which in turn is likely to facilitate creative problem solving. Knowledge has also been recognized as the essence in convergence: in order to synthesize ideas and develop them into something useable, sufficient knowledge is necessary (Tan, 2015). Such convergence is facilitated in the interaction with others (ibid.). To share knowledge between domains, the symbolic system used within a domain needs to be made available for people not accustomed with it (Csikszentmihalyi 1996). First when knowledge is understood by people and used in a purposeful way, it becomes valuable to the creative process (Basadur & Gelade, 2006). Hence, the availability of informations will not suffice if this information is not understood and incorporated by the designers in the design process (Basadur & Gelade, 2006).

31


32

Summary of results

4 SUMMARY OF RESULTS

4.1

House product platform

Paper I: Advancement of platform development in industrialised building In this paper, the development process of a house platform was studied, ranging from the first conceptual idea to finished construction drawings. The initial purpose of the paper was to increase knowledge of how to systematize platform development processes in a house building context in order to better control goal fulfilment. The design team’s task was to reuse experiences from previous projects and implement them in a house platform that would allow sufficient cost efficiency to build rental apartments in suburbs or small towns. This task specification was given orally to them by the company management. Cost efficiency was discussed in terms of efficient production, maintenance free solutions, short lead-times and optimized floor plans. The building should also be easy to place on different kinds of building plots, pleasant to live in and have a sense of quality. The design team consisted of a project leader who owned the responsibility of the development project, an installation coordinator, two site managers, an architect and a building engineer. They performed the product platform development project in parallel with their regular work tasks. Their design process was rather straight forward. At the first meeting they decided on the general appearance of the building: that it should be a space efficient tower block, that it should have maximum eight floors, what structural system to use, and that installations should be placed as close to the stairwell as possible. The architect was in charge of developing suggestions for the apartment layouts, and the other team members provided feedback from their respective perspectives. During their meetings, the discussions were open for everyone to have a say. The architect listened to the

33


feedback, went back to the drawing board, and presented a refined suggestion at the next meeting. Design decisions were taken in collaboration between everyone in the design team. No structured methods were used in the process except from a rough time plan with frequent meetings, managed by the project owner. Some ideas caused iterations in the process, e.g. the idea of a maximum expansion zone that would allow the building to be easy to place but still enable a variety of apartment layouts. The main concept of the building, which was decided at the first meeting, remained the same though, throughout the process. When the architectural design reached a level where everyone was satisfied with the results, the design process continued through design development (focusing on one of the combinations of apartment layouts). During this process, the design team continued their close collaboration, and additional consultants were added to finalize e.g. the structural design and electricity. The variability of the platform was documented and visualizations were made to support marketing. This documentation progressed alongside the first built projects, basing the documentation of the platform on what was found necessary when selling and producing the houses. 4.2

Standardized design development

Paper II: Design Management using Knowledge Innovation and Visual Planning The standardized design development process of one of Sweden’s leading industrialized house builders was studied in this paper. Their design process consists of design development and production planning, where externally produced conceptual drawings are synchronized with the company’s product platform constraints. The aim of the paper was to explain how design breakdown had been used to enable a Lean-approach to the design process. By breaking down the design process into well-specified activities, the company wanted to develop a support structure that would increase knowledge sharing within the design team. The support structure was also meant to enable reliable time management and increased design pace to match design process outcomes to factory production capacity. By standardizing design process activities, the process could be continuously refined using built up knowledge. Each design activity is specified using standard operations sheets, defining what to do, how to do it, responsible designer and when and where to deliver the output. The aim of the standardization was to be able to perform each design task in the same way every time, regardless of project, and regardless of who was performing the task. A prerequisite for this structure is the company’s product platform, where building system, technical solutions, and production aspects

34

Summary of results

are predefined. Deviations from the platform are not encouraged in the design process (however, the platform is continuously refined based on ideas that emerge in projects). By using a design process structure, throughput time in design development was reduced and experienced as more predictable. The design process participants participated in defining the standardized structure, thus integrating their understanding of their own work tasks into the task specifications. Clear task specifications has increased the design process participants’ understanding of each other’s work, and enabled reliable time management. The standardized design process is visualized on magnetic boards; one for each active project. Each activity has a magnetic marker which is used for time scheduling of the design work. Though there is a master schedule that specifies when each activity should be scheduled, the variety of projects in terms of type, size and complexity calls for flexibility regarding the length of the design process. This is adjusted manually by responsible design managers. The design team and representatives from the purchasing and factory departments meet around the magnetic boards each morning to carry out a daily review of the projects. The design process structure has shown to be good support when discussing design progression and sharing perspectives on the design. However, the external architects’ design processes were not yet incorporated as one of these perspectives, at the time of the study. 4.3

Design automation platforms

Paper III: Modularization Based on Commonalities in House-Building Requirements The company studied in this paper develops module-based product platforms based on customers’ product portfolios. Hence, the main part of systems design is performed previous to the product platform development process. The final definition of what systems to include in the platform is however decided in the platform development process. The platforms are used for design automation of e.g. house sub-systems, delivering construction drawings as the final result. The initial purpose of the study was to explore the design approach in order to elucidate how requirements are found and managed in the process. The case company strives towards implementing a structured requirement management approach. Due to the variety of products included in their platforms, both flexibility and structure has however shown important. The design process was possible to describe in a sequential manner, with emphasis on the frequent iterations between the process steps. The design process starts by analyzing the customer’s requirements on the design automation tool itself and on the products that are to be defined within the tool. These requirements are

35


continuously updated, since it has shown difficult to clearly understand how some requirements will influence the final result before this has been visualized. Hence, the designers (one designer has the main responsibility in each project) have continuous communication with the customer. The designers also analyze the customer’s portfolio products to grasp their scope and find commonalities between them. This is then used to explore and synthesize different design possibilities in order to eventually arrive at a few module definitions with built-in variability that satisfies the intended scope and fulfills the requirements. The aim is not to find new solutions, but to reuse the knowledge that is built up in the customers’ portfolios. This was experienced as an intellectual challenge by the designers. Specific predefined methods are used throughout the process to ensure that all requirements are found, understood and managed in such a way that their successful fulfillment is ensured. Much of the success of the modularization process is however dependent on the knowledge of the designers who interpret and analyze the scope of portfolio products and manage the design process the way that they see most fit. There is a design process description meant to support the designers in what to do when, but this is only partially used in the projects since each project is different from the others. Continuous customer dialogue is used to ensure that the designers have understood the design intention and the products in a satisfactory way, reducing the risk for misunderstandings. However, a production perspective was at the time of the study lacking in this dialogue since this perspective was not part of the clients’ organizations. The production perspective is instead incorporated as continuous improvement of the platform, based on feedback from platform use. 4.4

Architectural design

Paper IV: Arkitekters syn på kravhantering vid byggnadsutformning: Resultatredovisning (English: Architects’ views on how requirements are managed when designing buildings: A result presentation) This paper presents the results from an interview study with five architects with varying degree of experience from designing buildings for industrialized house building. The interviews focused on how requirements and constraints are managed in design processes, particularly in settings where contractors’ predefined building systems are to be used. The initial purpose of the study was to elucidate the architects’ reflections on their day-to-day work. Though concepts such as requirement management, design automation, configuration and parameterization were used in the interview questions, which implies

36

Summary of results

some kind of management strategy and thus some kind of structure, the architects described their design process as mainly intuitive. The design task can vary from well-defined and straight forward in some projects to ill-defined and open-ended in other projects, depending on the customer. The architects seem to prefer a client with a strong ambition rather than being given free rein. If a task is well-specified, they experience it easier to arrive at suitable solutions. It is also common that requirements on the house can evolve during the process, based on changed conditions or new ideas that emerge by the architect, client or other stakeholders. Unquantifiable qualities such as aesthetics, symbolic value and functional space considerations are always included in the design task. The architects described that there is always an overarching vision that is important to not lose sight of. This is the main role of the architect, to see the big picture regarding the design result. Some of the constraints and requirements that are posted on the building can sometimes be contradicting or difficult to fulfill without losing other important qualities of the design. This makes the architects experience the design task as complex and challenging. The architects manage these conflicts through intuition, experience and communication with the customer. If constraints from contractor platforms are introduced late in the design process, the architects have experienced that this often leads to unnecessary compromises and reduced architectural quality of the final design. Knowledge of production constraints, economic aspects, site conditions, and customer requests should be available as soon as possible in the design process in order to properly weigh different aspects of the design against these perspectives. The architects have experienced it important to be able to discuss reasons behind constraints and in this way be better informed to make decisions regarding the design, in some cases pushing the boundaries of the platforms. They describe the best way to share knowledge to be through communication, suggesting close collaboration between stakeholders as a means to improve knowledge integration in the design process. Written documents can serve as support, but should not replace dialogue, according to the interviewed architects. Even though many architecture offices have checklists that the architects can use, these are seldom used during the design process since they are perceived to impede the creative flow. Instead they are used before delivering the design to the customer to make sure that all important aspects have been considered. The main structure of the design process is the time plan, which is based on the collaboration with the other stakeholders. The design process is described to not be systematic. One of the architects compares it to solving a puzzle. Its success depends on the architect’s ability to analyze, understand the design situation and make good judgments of how to

37


prioritize different qualities. It happens that a design suggestion is discharged a while into the design process. One of the architects describe that this is seldom received well by the engineers working in a project since they are used to having their design problem well-specified in advance. The architect reflect on this issue as a difference between the design process in conceptual design, where retakes on the design are common, and on the design process in systems or detailed design, where the design task is usually more straight forward. Despite such retakes, the architects describe their process as rather linear, though including an open minded approach to testing new design suggestions, regardless of where in the design process they happen to be. Some projects also lead to more straight forwards design processes than others. When the end result is expected to be exclusive, the design process can also allow some more exploration.

38

Analysis

5 ANALYSIS

Figure 5.1. Comparison of the four studied design approaches.

39


In the following sections, the four design approaches are analyzed based on the theoretical framework presented in this thesis. The analysis is qualitative in character, focusing on reflection rather than quantification of the creativity influencing factors presented in Table 1. A visualization of the design approach features is presented in Figure 5.1, based on the qualitative analyses presented below. 5.1

Design approach to develop a house product platform

The design task given to the platform development team did not contain any clear requirements for novelty or originality, which is one of the facets of the creativity construct (e.g. Batey, 2012). They were expected to reuse experiences from previous projects. In this sense, the design task can be regarded as rather algorithmic (Amabile, 1996), since the development team knew already from the start what kind of building they should design and what kind of structural system and installations they should use. The open-ended parameters were mainly that the product should be sellable, easy to place on different types of building plots, have pleasant living spaces, and give a sense of quality. These parameters were also managed by synthesizing experiences from previous projects. It is therefore not likely that the design task itself contributed to facilitate creativity, since it was well pointed out that new creative solutions were not sought for. Instead, focus of the task was to achieve value for the organization, the client and the end users by cost efficiency, using well-tested previous solutions. The design process can be described as highly convergent, focusing on one idea which was gradually refined in more and more detail (Pahl et al. 2007; Runco, 2007). The process was in the appended paper described using an engineering design process theory as support, and the linearity of the process was quite evident. Even though the platform development team also managed artistic considerations such as appearance, usability and marketing possibilities (Eder, 2013), their design process can be described as towards the engineering design side of the polarized scale in Table 1, based on its converging character. However, the strive to structure and manage the design process, as described by Le Masson et al. (2011), could not be found in this platform development process. They did not employ any prescriptive engineering design methods (Cross & Roozenburg, 1992). Their approach was highly autonomous and could possibly be described by Le Masson et al.’s (2011) expression of autonomous creators working in creative teams. Such autonomy is likely to facilitate creativity if the designers have the experience and personality types to benefit from such conditions (Chang et al., 2012; Oldham & Cummings, 1996).

40

Analysis

In addition, the close collaboration between designers of different knowledge backgrounds, working in what Caniëls et al. (2014) describe as ‘safe environments’, is likely to further improve creativity (Carlisle & Dean, 1999; Csikszentmihalyi 1996; Gilson et al., 2005). The design approach seem to facilitate creativity, given the degree of autonomy and knowledge sharing. However, since creativity was not required nor encouraged, the algorithmic task in combination with the convergent process, which was enabled thanks to the high degree of knowledge sharing, seem likely to reduce this creative potential. 5.2

Standardized design development approach

The design task managed by the designers in the standardized design development approach aimed at finding ways to synchronize the conceptual design, which is provided to them by external architects, with the predefined platform constraints. Focus is on physical performance since artistic considerations have already been dealt with by the external architects. This is in accordance with Cross & Roozenburg’s (1992) description of an engineering design task. Based on the company’s design process standardization, it is clear that the design task itself can be regarded as distinctive (Lawson, 2005) and well-defined (Cross & Roozenburg, 1992), with little originality (Howard et al., 2008) (and hence little creativity) expected from the designers in individual projects. Originality can instead be incorporated in the continuous refinement of the platform. From a creativity perspective, the design task can be described as mainly algorithmic (Amabile, 1996) since it is well-specified already from the beginning. However, since each project is unique in its architectural design, some flexibility is required from the designers and the design task can be regarded to have some heuristic characteristics as well. In addition, the complexity of the task suggests that it is likely to be perceived as intellectually challenging which, according to Hunter et al. (2007) and Shalley et al. (2004), can contribute to creativity. This suggests that even though the design task will be towards the algorithmic side of the scale in Figure 5.1, and the designers are not expected to achieve originality, the design task can still provide a sense of challenge that can increase intrinsic motivation and thus provide some potential for creativity. The well-defined design process structure with clear time management can be compared to the linear character of the engineering design processes described by Cross & Roozenburg (1992), Pahl et al. (2007), and Wikberg-Nilsson et al. (2015). The case company has also adopted a prescriptive design process standardization, which is in accordance with how Cross & Roozenburg (1992)

41


view the engineering design consensus model. Though some divergence is likely to be required during the design process in order to find the best fit between the architectural design and the company’s product platform, the process can still be described as mainly converging, see Figure 5.1. This is based on the descriptions of Le Masson et al. (2011) and Cross & Roozenburg (1992), that the design process proceeds from the general to the particular and since subproblems are clearly defined in the design process standardization. Based on the distinction that divergence and convergence are needed in different stages of a design process (Caniëls et al., 2014), the converging character shown in this design approach can be seen as part of an implementation process (Howard et al., 2008) of the initial conceptual design. Flexibility, task-orientation and result-orientation are according to Caniëls et al. (2014) important during idea implementation. The converging character is also in accordance with the algorithmic character of the task. These should be in balance according to Runco (2007). When it comes to the design organization, it is clear that there is a distinct imposed structure that the designers are meant to conform to.. Cross & Roozenburg (1992) argue though that autonomy could still be allowed in a design process even though such a structure is present. Given the detailed level of task specification that is used at the case company, autonomy can hardly be argued as facilitated in this design approach, especially considering Chang et al.’s (2012) definition of autonomy as the lack of pacing and definition of subtasks. Clear structure can however positively influence creativity if the design participants for example are novices, have low self-control (Chang et al., 2012), or low creative ability (Choi et al., 2009; Oldham & Cummings, 1996). However, if the designers have high creative ability and/or previous experience of similar tasks, the structure imposed by the design organization is likely to have negative correlations with creativity (based on the findings of Chang et al., 2012, Choi et al., 2009 and Oldham & Cummings, 1996). This discussion should be related to the creative intention of the design task. Since the aim of the design approach is not to achieve high levels of creativity, the structure imposed by the design organization could in this case be regarded as a means to achieve other benefits, such as reliability in time management. Such benefits were also described by the interviewed managers to have been achieved by the introduced structure. Whether or not the imposed structure should be regarded as standardization or routinization, given the definition by Choi et al. (2009), is difficult to tell from the available empirical data. Projects vary in scope and content, but the design tasks are meant to be executed in the same way each time, suggesting both standardization and routinization. Again, given that the aim of the design approach is not to achieve creativity,

42

Analysis

routinization might not negatively affect the value of the final solution. However, one should reflect upon how routinization is likely to influence individuals’ intrinsic motivation (Csikszentmihalyi, 1996) which might in extension affect their performance, productivity and job satisfaction (Forgeard & Kaufman, 2016). The well-defined process structure in the standardized design development approach has shown to facilitate knowledge sharing at the case company. The designers experience that their understanding of the design process and of each other’s work has been increased. Hence, knowledge from different domains have been made accessible to the designers by the use of a standardized design process, which can facilitate idea generation and thus creativity (Csikszentmihalyi 1996). This can also be regarded as a positive effect considering the importance of knowledge sharing when working with complex systems (Carlisle & Dean, 1999; Le Masson et al., 2011). The design process structure also provides support for the designers to know what information they need in order to finalize each subtask. By integrating different functions (e.g. structural, purchasing and production) in the design process structure, knowledge can be communicated and understood in a way that ensures its successful implementation in the design, which is decisive according to (Basadur & Gelade, 2006). The knowledge sharing has however yet to include the architects working in conceptual design in a satisfactory way, according to the case company representatives. 5.3

Design approach to develop design automation platforms

The design task that the designers manage when developing the design automation platforms can be regarded as an engineering design task since it is focused on engineering perspectives such as functional behavior and physical properties (see Cross & Roozenburg, 1992). However, the task has shown difficult to fully define at project initiation: it is difficult to predict how requirements will influence design results before this can be tested and visualized. Hence, the design task can be described as to some degree heuristic (Amabile, 1996): the problem definition is allowed to evolve as the design process progresses (Lawson, 2005). The fact that the designers are enabled to reflect on and take part in the problem definition is likely to positively influence creativity (Csikszentmihalyi, 1996; Mumford, 2000; Runco, 2007). However, the aim of the design process is not to achieve creative products (considering the originality aspect of creativity), but to synthesize the customer’s portfolio solutions into a module-based product platform that covers the scope of the customer’s projects. This transformation from unique

43


products to module-based platform definitions was experienced by the designers as an intellectual challenge, which suggests positive influence on creativity (Hunter et al., 2007; Shalley et al., 2004). The design process is focused on synthesizing previously defined solutions: understanding them, finding commonalities and combining them to reduce the variety of solutions. This process is clearly convergent, which facilitates implementation rather than idea generation (Caniëls et al., 2014). The case company implements a structured approach to requirement management to ensure traceability of design decisions. There is also a process structure that the designers are encouraged to conform to. However, the variety of products and the variety of scope of the projects has promoted flexibility in the approach. This seem coherent with the contradiction described by Gilson et al. (2005), that structure is used to manage complexity and ensure consistent operations, but flexibility is necessary to cope with uncertainties. Though the design process was shown possible to describe sequentially, the designers described that there are frequent iterations between the process steps. Their inherent knowledge and their increasing understanding of the products were the foundation of the process, rather than the defined process templates. The process template did, at the time of the study, provide support rather than rules to strictly follow. This suggests that the design organization imposes structure, but still enable some autonomy of how to incorporate this structure into each platform development project. The support structure is likely to facilitate creativity if the designers are inexperienced or have low creative ability or selfcontrol (Chang et al., 2012; Choi et al., 2009), since it might reduce the feeling of ambiguity (Chang et al., 2012). But since the designers are trusted to manage their own processes, autonomy is possible to experienced designers and designers who perform better in autonomous conditions. Hence, the structure imposed on the design process seem to potentially enable creativity. However, if the structure was to be imposed as strict rules to follow, this potential would likely be decreased. In the design approach to develop design automation platforms, knowledge is shared between case company employees (however, one designer has the main responsibility of defining the product) and between representatives. The communication between designers and customer representatives allow continuous feedback on the progression of the product specification. Though not being teamwork per se, as suggested by Carlisle & Dean (1999) and Gilson (2005), this continuous communication is likely to facilitate knowledge sharing. This was argued by the designers as essential in their design process, due to the variety of product that they work with. The knowledge sharing is

44

Analysis

however limited to representatives from the customer organization. This means that perspectives from e.g. production is seldom included in the process. These perspectives are instead fed back from platform use, as continuous improvement of the platform. Hence, creativity seems to be somewhat facilitated by the continuous sharing of knowledge between the case company and its customers. 5.4

Architectural design approach

The architectural design approach is described by the architects as mainly heuristic: the task is rarely well-specified at project initiation and the architects are often encouraged to bring their own knowledge and preferences into the design. The degree of creativity that is expected from the architects varies depending on project type and scope, as well as the client’s creative ambition. The task is artistic in the sense that it includes e.g. appearance and usability (Eder, 2013). The degree of freedom in the design task varies between different projects, depending on the client’s preferences. By having influence over the design task, the architect experience that it is possible to not only bring their own values into the design, but also to provide feedback and push the boundaries of the platforms used by clients or industrialized house builders, encouraging knowledge sharing. In addition, the design task is not perceived as static: based on the increasingly detailed understanding of the building (Lawson, 2005), clients, architects or engineers might request changes to the design. This can be due to changed conditions or new ideas and this is considered as something natural in most projects. Hence, this suggests a divergent character of the process. The task is experienced by the architects as complex and challenging, which according to Hunter et al. (2007) is likely to facilitate creativity. Though the architects are positive about the opportunity to influence the design task, they describe it easier to work in projects where the client posts more extensive requirements. If the task is rather well-specified, the architects experience it easier to arrive at satisfactory design solutions. This seems coherent with Csikszentmihalyi’s (1996) view that algorithmic tasks are easier to solve but are likely to result in less creativity. Though heuristic tasks are clearly positive for creativity (Amabile, 1996), the architects experienced well-specified tasks as less likely to end up having to cut architectural qualities due to e.g. economic or production perspectives that are incorporated later on in the design process. To decrease this risk, the architects suggest knowledge sharing in multi-functional teams throughout design. This is seldom achieved today, but the architects perceive it as an improvement possibility in the building

45


industry. Multi-functional teamwork is likely to facilitate creativity (Carlisle & Dean, 1999), making knowledge available through cooperation rather than documentation in checklists or templates. Hence, the information could be more easily communicated and incorporated as shared knowledge for the design process participants (Basadur & Gelade, 2006). Knowledge sharing in safe environments was also found by Caniëls et al. (2014) to positively influence idea generation. Today, knowledge is mainly shared in project meetings or through email conversations. The sub-optimization culture in the construction industry (Lidelöw et al. 2015) is perceived as negative by the architects: information that reaches the designer too late can cause untimely changes and delays in projects. The design process seems coherent with Lawson’s (2005) description of artistic design processes as being built up of iterations between analysis, synthesis and evaluation. Both convergence and divergence can be recognized in the architects’ process descriptions. Divergence is mainly recognized in their descriptions of how they generate design ideas and how they can work on one idea for a while just in order to later on discard it and start new. The converging parts are when design ideas are further developed, evaluated and communicated, similar to the description by Howard et al. (2008). The architects perceive the engineering design process in design development as considerably more straightforward than their own design process. How straight forward their own process is depend on the design task. The architects describe that when working on ‘simple’ projects, the design process is rather straight forward because the task is rather straight forward. If the client on the other hand has high expectations on architectural expression and there is sufficient room in the budget, the design process will be more explorative and divergent. Their own process is characterized by autonomy and individuality, as theorized by Lawson (2005). Imposed structure consists of a project time plan and sometimes checklists to make sure that all requirements have been considered. Other than that, no process standardizations were mentioned by the architects. This suggests autonomy and possibility to diverge in the design process, which is likely to facilitate creativity (Le Masson et al., 2011), especially for experienced architects and architects with high creative ability (Chang et al., 2012; Choi et al., 2009).

46

Discussion and concluding remarks

6 DISCUSSION AND CONCLUDING REMARKS

The studied design approaches display a variety of design approach features. The theoretical conceptualization, building on the work of Le Masson et al. (2011), enabled a visualization of these differences. First of all, it is clear that an architectural design approach is significantly towards the artistic side of the scale in Table 1. The project-based design approach that was studied at an industrialized house builder is significantly towards the engineering side of the scale. This difference can be related to when in the design process these design approaches are focused. The architectural design approach is focused on conceptual design where ideation is central. Hence, it is not surprising that this approach facilitates flexibility, autonomy and, given the theoretical framework, creativity. It should however be noted that architects manage both the general and the detailed in their processes, including both divergence and convergence. But their main task is to develop a conceptual idea for the building, which will be further refined in the design development stage. Hence, the risk of the architectural design deviating from the initial intention, as identified by Kiviniemi (2005), might from a creativity perspective be considered as natural: the architects are allowed to influence the design task as new ideas emerge in the design process. The standardized design development approach on the other hand, is focused on later stages of a house design process where implementation is in focus. The case company placed high priority on reliability regarding time management and delivery of design results in order to facilitate production planning. This motivated the introduction of a standardized structure, which was perceived by the interviewed project managers to lead to the expected benefits. Though the designers’ tasks in this case can be regarded as complex, they are not expected to generate original (and hence not creative) solutions.

47


From a creativity perspective, the differences between these two approaches seem logic: the creative intention in architectural design is high (but varying depending on project) and the creative intention in the studied design development process is low (the design task is centered on finding a fit between platform predefinitions and externally produced architectural drawings, hence not aiming for novelty). Accordingly, creativity seem to not be facilitated to the same extent in the standardized design development approach as in the architectural design approach. Interestingly though, knowledge sharing is one design approach feature where the engineering side of the scale facilitates creativity more than the artistic side of the scale. This was also something that the interviewed architects reflected upon; they believed that a higher degree of teamwork would improve the chances of arriving at a highly successful end product (the end product can however be successful without being creative). Though teamwork can facilitate knowledge sharing and an increased understanding of different knowledge domains which, according to Csikszentmihalyi (1996) among others, is positive for creativity, it is also possible that teamwork could reduce the sense of autonomy towards the task since designers need to consider each other’s preferences as well. The effects of teamwork is therefore interesting to continue exploring, perhaps including factors such as co-worker relationships, feedback, and leadership. The two platform development approaches seem to both include a mix between artistic and engineering features. They are both aiming at developing products of market value, meant to be sold repeatedly. One could say that the design organization is fully in charge of the entire design process of the platform. This can be compared to the two studied project-based design approaches where the first part of the design process is performed by architects while the second part of the design process is performed within the industrialized house builder’s organization. This can be reflected back to the suboptimization in the industry as discussed by Lidelöw et al. (2015). By ‘owning’ the entire design process when developing platform products, the designers need to manage both ideation and implementation. Hence, the mix between artistic and engineering features is not surprising. What differentiates the two approaches from each other is mainly the degree of autonomy and the degree of teamwork. The case company who develops the design automation platforms develops new platforms on a regular basis, based on customer orders. Their process structure can therefore be reapplied between projects and continuously updated based on built up experiences (i.e. it serves as a process platform according to the definition by Robertson & Ulrich, (1998)). This motivates its existence, since it facilitates knowledge build up in the organization and reliability regarding design outcomes. It also facilitates knowledge sharing, since the designers will

48

Discussion and concluding remarks

have support regarding when and how to ask for what information from the client organization. The house product platform on the other hand was developed as a one time event. This explains the high degree of autonomy in how to go about the process. Thanks to the multi-functional composition of the design team, knowledge from multiple perspectives could easily be integrated in the product definition, facilitating convergence in the process and potentially also facilitating creativity. Neither of the two platform development processes had creativity as an initial intention. Accordingly, they also seem quite moderate in how they facilitate creativity. However, it seems reasonable to assume that if the creative intention of the house platform would have been higher (i.e. if the intention would have been to find novel solutions instead of synthesising previous solutions) the design approach could have included clear divergence in addition to an autonomous approach characterised by teamwork and knowledge sharing. Hence, this design approach could have strongly facilitated creativity. In relation to this reflection, a reflection by Choi et al. (2009, p. 342) is worth repeating: “Intuitively, it stands to reason that if a job does not require creativity as part of its performance, employees may not perform any creative behaviour even in the most creativity-promoting work context.” In summary, it seems as though the creative intention is related to how design is approached. From an industry perspective, this is something that should be reflected upon when deciding how to approach a design task. By understanding that different design approaches will influence the potential for creativity in different ways, which is something that this licentiate thesis has contributed in pointing out, a first step has been taken towards understanding why architects and industrialized house builders approach design differently. This is a relevant first reflection regarding the tension between architectural design practice and design practices in industrialized house building. They seem to approach design differently because they prioritize different values. Architects work with ideation where artistic considerations and creativity are considered giving value. Industrialized house builders work with implementation and engineering perspectives, hence valuing reliability and control. As a theoretical contribution, the theoretical framework presented in this thesis provided a lens that made it possible to distinguish differences in design approaches. It could be used to reflect on creativity implications in both a platform development context and in project-based design. Hence, the

49


synthesis of the theoretical framework is in itself a contribution to industrialized house building research. By further refining this framework and validating it through more empirical investigations, we could become closer to a richer understanding of the tension between architectural design practice and design practices in industrialized house building.

50

Future research

7 FUTURE RESEARCH

The study presented in this licentiate thesis address some of the features of design approaches that have relevant influence on creativity. However, it should be recognized that the creativity perspectives presented herein does not cover all creativity relevant aspects of design approaches in industrialized house building. Other aspects related to the social environment, such as coworker relationships, resource availability or external motivators such as rewards (Anderson et al., 2014) etc. are aspects that would also be relevant to study in order to additionally increase our understanding of the topic. In addition, the framework of design approach features would also be interesting to apply in different industrialized house building settings, exploring how different levels of pre-engineering (based on Bonev et al., 2015 and Lidelöw & Olofsson, 2016) or different types of product platforms (based on Jansson et al., 2014) might entail different approaches to design. This licentiate thesis does however provide an initial framework that has enabled relevant reflections. The framework could be further tested and refined to continue researching differences in design approaches, their motivations and their implications. In this thesis, the research was focused on design approaches on each side of the interface, rather than the actual interface itself. This was a necessary first step since one must first understand each other before one can find a common path forward. As a next step, efforts to bridge the differences between architectural design practice and design practice at industrialized house builders could be the focus of investigation, based on the increased understanding of how and when to facilitate creativity. An example could be to look into how architects’ prerequisites are affected by collaborating with industrialized house builders who use building systems predefined in product

51


platforms. Relevant questions are how the architects’ potential to create varied architecture is affected by standardized building systems, and how their everyday practice is affected by the changed project settings and constraints entailing industrialization. Another example could be to explore possibilities to develop a design process support structure that maximizes creative potential during conceptual design while still enabling efficient incorporation of product platforms, clear deliveries and knowledge sharing between stakeholders in the process. On the creativity topic, one interesting possibility is to further elucidate platform approaches from a design constraint perspective. It seems as though standardization facilitates knowledge sharing, which has been recognized by multiple researchers as positive for creativity (e.g. Csikszentmihalyi 1996; Gilson et al., 2005). However, by constraining the structural system of the building, the design task becomes more well-defined, but also more complex. This suggest both negative and positive synergies with creativity. Obviously, there are multiple creativity influencing factors that are likely to be affected by the incorporation of predefined platforms in the design process. These would be relevant to further explore. By better understanding how design creativity is increased or reduced in an industrialized house building context, awareness can be raised regarding how distinctiveness could be facilitated and balanced with commonality and rationality to the desired degree, which according to Bonev et al. (2015) and Jansson et al. (2014) is essential in industrialized house building. Further, the analysis presented in this thesis is based on a qualitative interpretation. It would be interesting to develop the theoretical framework to enable also quantitative measures, which would allow better reliability when comparing design approaches and their implications on creativity. An interesting example is that researchers have found that e.g. autonomy seem to have less effect on creativity than other creativity enablers (Hunter et al., 2007). This is an interesting reflection since managers should be allowed to weigh investments in creativity facilitators against their expected outcomes. Another example is knowledge sharing, where an increased understanding of different knowledge domains is beneficial for creativity and can be facilitated by multifunctional teamwork. When discussing teamwork and creativity, creativity influencers such as co-worker relationships and leadership also seem relevant to the discussion. This is something that could be further added to the theoretical framework of this thesis. To develop and validate a framework that would allow quantitative measures is however likely to require quite some effort. Such research should therefore be carefully planned, weighing expected outcomes from the research with the effort required to successfully conduct it.

52

Bibliography

Bibliography

Anderson, N., Potočnik, K. & Zhou, J. (2014) Innovation and Creativity in Organizations: A State-of-the-Science Review, Prospective Commentary, and Guiding Framework, Journal of Management, 40(5), 1297-1333 Amabile, T.M. (1983) The Social Psychology of Creativity. New York: SpringerVerlag Amabile, T. M. (1996). Creativity in context: Update to" the social psychology of creativity", Boulder: Westview press Amabile, T.M. (1997) Motivating Creativity in Organizations: On Doing What You Love and Loving What You Do, California Management Review, 40(1), 3958 Apleberger, L., Jonsson, R. & Åhman, P. (2007) Byggandets industrialisering: Nulägesbeskrivning, Göteborg: Sveriges Byggindustrier. Basadur, M. & Gelade, G.A. (2006) The Role of Knowledge Management in the Innovation Process, Creativity and Innovation Management, 15(1), 45-62 Batey, M. (2012) The Measurement of Creativity: From Definitional Consensus to the Introduction of a New Heuristic Framework, Creativity Research Journal, 24(1), 55-65 Bergendahl, M., Magnusson, M. & Björk, J. (2015) Ideation High Performers: A Study of Motivational Factors, Creativity Research Journal, 27(4), 361-368 Biskjaer, M.M., Dalsgaard, P. & Halskov, K. (2014) A Constraint-Based Understanding of Design Spaces, DIS 2014, June 21-25, Vancouver, Canada

53


Bonev, M., Wörösch, M. & Hvam, L. (2015) Utilizing platforms in industrialized construction: A case study of a precast manufacturer, Construction Innovation, 15(1), 84-106 Caniëls, M.C.J., De Stobbeleir, K. & De Clippeleer, I. (2014) The Antecedents of Creativity Revisited: A Process Perspective, Creativity and Innovation Management, 23(2), 96-110 Caniëls, M.C.J. & Rietzschel, E.F. (2015) Organizing Creativity: Creativity and Innovation under Constraints, Creativity and Innovation Management, 24(2), 184-196 Carlisle, Y. & Dean, A. (1999) Design As Knowledge Integration Capability, Creativity and Innovation Management, 8(2), 112-121 Choi, J.N., Anderson, T.A. & Veillette, A. (2009) Contextual Inhibitors of Employee Creativity in Organizations: The Insulating Role of Creative Ability, Group & Organization Management, 34(3), 330-357 Cross, N. & Roozenburg, N. (1992) Modelling the Design Process in Engineering and in Architecture, Journal of Engineering Design, 3(4), 325-337 Csikszentmihalyi, M. (1996), Creativity - Flow and the Psychology of Discovery and Invention, New York: Harpercollins Publishers Eder, W.E. (2013) Engineering Design vs. Artistic Design: Some Educational Consequences, US-China Education Review A, 3(4), 259-280 Flick, U. (2007) Managing Quality in Qualitative Research, London: SAGE Publications Forgeard, M.J.C. & Kaufman, J.C. (2016) Who Cares About Imagination, Creativity, and Innovation, and Why? A Review, Psychology of Aesthetics, Creativity, and the Arts, 10(3), 250-269 Gibbs, G.R. (2007) Analyzing qualitative data (1. ed.), Thousand Oaks.: Sage Publications. Gillham, B. (2000) Case Study Research Methods, London: Continuum Gillham, B. (2005/2008) Forskningsintervjun: Tekniker och genomförande (E. Jamison Gromark, Trans.), Lund: Studentlitteratur, (Original work published 2005)

54

Bibliography

Gilson, L.L., Shalley, C.E. & Ruddy, T.M. (2005) Creativity and standardization: Complementary or conflicting drivers of team effectiveness?, Academy of Management Journal, 48(3), 521-531 Green, L.N. & Bonollo, E. (2002) The Development of a Suite of Design Methods Appropriate for Teaching Product Design, Global Journal of Engineering Education, 6(1), 45-51 Hansson, B., Olander, S., Landin, A., Aulin, R. & Persson, U. (2015) Byggledning – Projektering, Lund: Studentlitteratur Hennessey, B.A. (2003) The Social Psychology of Creativity, Scandinavian Journal of Educational Research, 47(3), 253-271 Hirst, G., van Knippenberg, D. & Zhou, J. (2009) A cross-level perspective on employee creativity: Goal orientation, team learning behavior, and individual creativity, Academy of Management Journal, 52(2), 280-293 Hon, A.H.Y. (2012) Shaping Environments Conductive to Creativity: The Role of Intrinsic Motivation, Cornell Hospitality Quarterly, 53(1), 53-64 Howard, T.J., Culley, S.J. & Dekoninck, E. (2008) Describing the creative design process by the integration of engineering design and cognitive psychology literature, Design Studies, 29(2008), 160-180 Hunter, S.T., Bedell, K.E. & Mumford, M.D. (2007) Climate for Creativity: A Quantitative Review, Creativity Research Journal, 19(1), 69-90 Håkansson, H. & Ingemansson, M. (2013) Industrial renewal within the construction network, Construction Management and Economics, 31, 40-61 Jansson, G., Johnsson, H. & Engström, D. (2014) Platform use in systems building, Construction Management and Economics, 32(1-2), 70-82 Javid, A. (2014) Creativity in Architecture Design Eduction: Design as Puzzle Solving, The International Journal of Design Education, 8(2), 11-21 Johnsson, H. (2013) Production strategies for pre-engineering in housebuilding: exploring product development platforms, Construction Management and Economics, 31(9), 941-958 Kandler, C., Riemann, R., Angleitner, A., Spinath, F.M., Borkenau, P. & Penke, L. (2016) The Nature of Creativity: The Roles of Genetic Factors, Personality Traits, Cognitive Abilities, and Environmental Sources, Journal of Personality and Social Psychology, 111(2), 230-249

55


Kaufman, J.C. & Sternberg, R.J. (Eds.) (2010) The Cambridge Handbook of Creativity, New York: Cambridge University Press Kiviniemi, A. (2005) Requirements Management Interface to Building Product Models (Doctoral dissertation, Stanford University, Department of Civil and Environmental Engineering) Langley, A. (1999) Strategies for theorizing from process data, The Academy of Management Review, 24(4), 691-710 Lawson, B. (2005) How Designers Think - The Design Process Demystified (4 ed.), Oxford: Architectural Press Le Masson, P., Hatchuel, A. & Weil, B. (2011) The Interplay between Creativity Issues and Design Theories: A New Perspective for Design Management Studies?, Creativity and Innovation Management, 20(4), 217-237 Lessing, J. (2006) Industrialised House-Building: Concept and Processes (Licentiate dissertation, Lund University, Department of Construction Sciences) Lessing, J., Stehn, L. & Ekholm, A. (2015) Industrialised house-building: development and conceptual orientation of the field, Construction Innovation, 15(3), 378 - 399 Liang, C., Chang, C.-C. & Hsu, Y. (2013) Differential Effects of Personality Traits and Environmental Predictors on Reproductive and Creative Imagination, The Journal of Creative Behavior, 48(4), 237-253 Lidelöw, H. & Olofsson, T. (2016) The Structure and Predefinition of the Industrialized Construction Value Chain, Proceedings of the 2016 International Conference on Construction and Real Estate Management, Edmonton, Canada, September 29 – October 1 Lidelöw, H., Stehn, L., Lessing, J. & Engström, D. (2015) Industriellt byggande, Lund: Studentlitteratur Merriam, S.B. (2009) Qualitative research: a guide to design and implementation (2. ed.) San Francisco: Jossey-Bass Mumford, M.D. (2000) Managing creative people: Strategies and tactics for innovation, Human Resource Management Review, 10(3), 313-351 Mumford, M.D. (2003) Where Have We Been, Where Are We Going? Taking Stock in Creativity Research, Creativity Research Journal, 15(2), 107-120

56

Bibliography

Mumford, M.D. (Ed.) (2012) Handbook of Organizational Creativity, London: Elsevier Oldham, G.R. & Cummings, A. (1996) Employee Creativity: Personal and Contextual Factors at Work, Academy of Management Journal, 39(3), 607-634 Onarheim, B. & Biskjaer, M.M. (2013) An Introduction to 'Creativity Constraints', Innovating in Global Markets: Challenges for Sustainable Growth, Proceedings, Lappeenranta University of Technology Press Pahl, G., Beitz, W., Feldhusen, J. & Grote, K.-H. (2007) Engineering design: a systematic approach, 3 Ed., Berlin: Springer Peponis, J., Bafna, S., Dahabreh, S.M. & Dogan, F. (2015) Configurational meaning and conceptual shifts in design, The Journal of Architecture, 20(2), 215-243 Robertson, D. & Ulrich, K. (1998) Planning for product platforms, Sloan Management Review, 39(4), 19-31 Runco, M.A. (Ed.) (1997) The Creativity Research Handbook, Cresskill, N.J.: Hampton press Runco, M.A. (2007) Creativity: Theories and Themes: Research, Development and Practise, Amsterdam: Elsevier Academic Press Runco, M.A. & Jaeger, (2012) The Standard Definition of Creativity, Creativity Research Journal, 24(1), 92-96 Sawyer, R.K. (2011) The Cognitive Neuroscience of Creativity: A Critical Review, Creativity Research Journal, 23(2), 137-154 Sawyer, R.K. (2012) Explaining Creativity: The Science of Human Innovation, Ed. 2, New York: Oxford University Press Shalley, C.E. & Gilson, T.C. (2004) What leaders need to know: A review of social and contextual factors that can foster or hinder creativity, The Leadership Quarterly, 15(2004), 33-53 Shalley, C.E., Gilson, L.L. & Blum, T.C. (2009) Interactive Effects of Growth Need Strength, Work Context, and Job Complexity on Self-Reported Creative Performance, Academy of Management Journal, 52(3), 489-505 Shalley, C.E., Zhou, J. & Oldham, G.R. (2004) The Effects of Personal and Contextual Characteristics on Creativity: Where Should We Go from Here?, Journal of Management, 30(6), 933-958

57


Sternberg, R.J. (Ed.) (1999) Handbook of Creativity, Cambridge: Cambridge University Press Stockholms Byggmästareförening Byggmästareföreningen

(2014)

Ett

hus

blir

till,

Stockholm:

Tan, A.-G. (2015) Convergent Creativity: From Arthur Cropley (1935-) Onwards, Creativity Research Journal, 27(3), 271-280 Viklund, E. (2015) Product Development of Industrialised Building Concepts: Exploring the Possibility of Utilizing Methods from the Manufacturing Industry (Master thesis, Engineering Technology: Architecture, Luleå University of Technology) Wikberg-Nilsson, Å., Ericson, Å. & Törlind, P. (2015) Design: process och metod, Lund: Studentlitteratur Wikner, J. & Rudberg, M. (2005) Integrating production and engineering perspectives on the customer order decoupling point, International Journal of Operations & Production Management, 25(7), 623-641 Winch, G.M. (2003) Models of manufacturing and the construction process: the genesis of re-engineering construction, Building Research & Information, 31(2), 107-118 Yin, R. K. (2011) Qualitative Research from Start to Finish, New York: The Guilford Press

58

Paper I

Paper I

ADVANCEMENT OF PLATFORM DEVELOPMENT IN INDUSTRIALISED BUILDING Author’s manuscript Gustav Jansson and Emma Viklund 8th Nordic Conference on Construction Economics and Organization Published in Procedia Economics and Finance 21 (2015), pp. 461 – 468 Published as an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) https://doi.org/10.1016/S2212-5671(15)00200-2

Abstract Demand for productivity in house-building is today causing changes of work methods in the building industry, for example by the utilization of house-building platforms. This requires development processes separated from those of individual house products. The aim of this study is to examine how platform development processes in the building industry are carried out and how product development theories fit to the studied context. A qualitative case study analysis shows how a house-building platform can be developed in a sequential manner with product development theories that could capture values from the construction company, property owner and from end user.

Introduction Industrialised construction is often compared to the manufacturing industry, where focus has shifted from massproduction to a more market-oriented development process of new products (Johannesson et al., 2013). But there are major differences between the building industry and the manufacturing industry in the description of using predefinitions as support for efficient and product oriented processes (Gibb, 2001; Lessing, 2006). The organisation surrounding a building project is often decentralised, which leads to a proliferation in methods and processes (Apleberger, et. al., 2007). When the building projects change from project orientation towards industrialisation with predefinitions, the methods for developing house-building platforms also need to change into one part that focuses on product development, and one part that is linked to the project specific detailing process (ibid.).

1

Advancement of platform development in industrialised building

Systematised theories of product development methods are described in the manufacturing industry but are lacking in the building industry. The degree of predefinition of a product, or in this case a house-building platform, is based on how much of the product that is defined at the customer order decoupling point (CODP) (Johnsson, 2013; Gosling & Naim, 2009). A dominant part of the building industry is in the engineer-to-order (ETO) context (Gosling & Naim, 2009), which means that the customer can take part in the product specification process before any components have been produced (Johnsson, 2013). Pre-engineering of building components, layouts and subprocesses before order point, enables the customer to choose and configure the design and functionality based on platform definitions (Olhager, 2003). This can be achieved within an ETO context (Johnsson, 2013). The value of different strategies becomes central for the operational platform, measured in terms of cost development, quality improvements and lead times to customer and market (Brege et al. 2013). Nevertheless is the house-building industry in a shift from project based construction to diversification of predefinitions in platform products in order to increase productivity (Jansson et al., 2014). To find efficient and reliable methods, knowledge about systematised development processes could be enhanced by analysing the house-building context to product development theories. The aim of this research is to examine how platform development processes in the building industry are carried out and how product development theories fit to the studied context. This presupposes an understanding of how different values in the house-building platform contribute in the development process. Descriptions of methods that can be used for structuring the development processes are lacking which forms the reason of this study. In order to fulfil the aim, the following research questions need to be answered: * How could structures for product development contribute in the development of house-building platforms? * How do different value perspectives affect platform development processes?

Frame of reference Platforms provide a way of gaining benefits of volume while at the same time being able to offer products that can be individually adapted to each customer (Robertson & Ulrich, 1998; Thuesen & Hvam, 2011; Williams, et al., 2007). The complexity of platform systematisation in industrialised house2

Paper I

building does not stem from the work of defining the physical building system but from identifying a balance between predefinitions that provide customer value for the client and economies of scale in the diversity of product features (Voordijk et al., 2006). It has been shown difficult to identify general structures for supporting creative work for diversity of house products. Therefore it is important to separate the development of house-building platforms from the development of individual house projects (Jansson, et al., 2014; Lessing, 2006). Jensen, et al. (2012) describe a house-building platform from three different views: engineering, customer and production. The engineering view is described by the technical platform in technical solutions and components (Lessing, 2006), the customer view by an architectural platform with architectural modules for technical, functional and aesthetic values (Wikberg & Ekholm, 2011), and the production view by the process platform with tools for supporting work processes (Lessing, 2006). Since each project in the building industry is unique regarding client demands and site conditions & regulations, it is difficult to have a fully defined house-building platform (Jansson, et al., 2014). Attempts (Jansson & Lundkvist, 2014; Bonev et al., 2014) have been made to evaluate how customer values, as for example functionality and variety, are kept in the development of house-building platforms. The results from these studies show that focus has been on reducing time and cost by using predefinition and routinized production rather than prioritizing customer values. When it comes to product development methods, the product oriented industry provides a pool of theories that are suitable to implement in different product development situations (Johannesson et al., 2013). Some methods are structured and some are based on intuition and experience. Studies have shown however that structured methods are positive for controlling the process and outcome of new product development (Graner & Missler-Behr, 2013). One way to structure a design process is by using an organised design approach (Pahl et al., 2007; Cross, 2008). This kind of approach can communicate a clear understanding of goals and work processes, and it supports interdisciplinary collaboration (Johannesson et al., 2013). Pahl et al. (2007) and Cross (2008) describe a general design approach from VDI-Richtlinie 2221 (1993) that is one example of how to structure a design process. The general subdivision of design activities in this approach makes it valuable for design projects regardless of industry branch (Cross, 2008). In this kind of approach, the output of each design stage should be used to evaluate the progression against the initial, and continuously updated, requirement specification (ibid.). Other approaches are for example axiomatic design, which focuses on product properties and their hierarchical relations in radical development (Suh, 2001), or quality function deployment (QFD) that uses a matrix structure for evolutionary refinement (Akao, 1990). 3


Design phases in construction projects are generally fragmented which leads to increased design time and cost, costly design changes, and sub-optimal design solutions due to lack of interdisciplinary communication (Evbuomwan & Anumba, 1997). Concurrent engineering (CE) can be used to improve performance in projects (Khalfan & Raja, 2012) and decrease the negative aspects of fragmentation by integrating production values in design activities (Evbuomwan & Anumba, 1997). This has been achieved successfully in the product oriented industry where benefits as reduced time-to-market, increased quality and reduced cost have been shown (Pahl et al., 2007). In order to improve development strategies for house-building platforms, structured development processes need to be set in relation to value perspectives and to the variation of different house-building contexts.

Research approach In Sweden, the construction and property development company NCC has developed a number of industrialised house-building platforms over the years. One of them is NCC Folkboende which was developed for the ETO context with the aim of selling Folkboende buildings like configurable products. NCC Folkboende was chosen for the study for its high level of pre-engineering and due to the fact that it was developed recently. Since the aim of this research is to examine how a house product development process was carried out in a real project, a case study approach is suitable (Yin, 2014). The research work included the platform development phase in a design-build organisation. When conducting a case study, the combination of different methods for gathering the necessary information enables a deep understanding of the phenomenon (ibid.), where archival data and observations at the office complement interviews. Respondent control was used to ensure the validity of the results (Merriam, 1994). The studied house-building platform The house-building platform of NCC Folkboende was developed at the local NCC office in Umeå, Sweden, based on methods and systems that had been developed and practiced over time in previous house-building projects. By providing a configurable house product, the aim was to minimise project specific and ad hoc solutions and to enable fast lead-times from project initiation to completion. This led to a high level of pre-engineering of the architectural, technical and process platform. Customisation is practiced by layout configurations, Fig. 1, and project specific solutions for the entrance floor.

4

Paper I

Figure 1. Architectural modularity of NCC Folkboende. The figure is based on the product variants of the Folkboende house product. The technical platform contains technical solutions ranging from predefined details to overall building system. The building variants are created by combinations of apartment types with different sizes, which are specified in the architectural platform. Figure 1 shows the base for configuration and the expansion zone which indicates the maximum footprint of the building. It is possible for the customer to choose the number of stories ranging between four and eight. All materials are fixed in the platform, but it is possible for the customer to choose colours. The process platform provides project support for planning and production, gathered in a web-based production support system which for example contains instruction videos for production. Case study analysis First the development process of the Folkboende house-building platform was mapped, identifying sequences of activities. When visualising the development stages, the architectural, technical and process platform predefinitions (A, T, and P in Figure 2) were identified in the mapped activities. Design iterations were also mapped to the sequence of stages (see arrows in Figure 2). The process was mapped through interviews with the participants of the development group, and by studying archival data and doing observations. Secondly, check-up questions were posted to increase objectivity. As a third step, the mapped development process was analysed against theories of product development from the manufacturing industry in order to discuss the development process of the house-building platform from a value perspective.

5


Development process of the Folkboende house-building platform The development group of NCC Folkboende consisted of a project leader, an installation coordinator, two site managers, an architect and a building engineer. A generally perceived need of increased production of rental apartments was the basis of the development project and the belief was that a house product could entail sufficient project economy to achieve this. The market focus was on rental apartments for the general population and the apartments were not meant to be built in prime locations but rather in the outskirts of larger Swedish cities or as densifying buildings in smaller cities. Small sized apartments, preferably bedsits and one-bedroom apartments, were the market demand. No specific market of property owners was intended. Since the development group was multifunctional, viewpoints of production, construction, installation and architecture could be developed in parallel. This was important in order to ensure that the platform fulfilled the values that the development group were aiming for. The main value was to reach cost efficiency, both for NCC, the property owner and the end user. For NCC, cost efficiency was mainly discussed in terms of efficient production, thus mostly affecting the technical and process platform. Cost efficiency for the property owner referred to maintenance free solutions and short lead-times, mainly affecting material choices and product definitions. Regarding the end user, cost efficiency was discussed related to optimised floor plans which would lead to low rents, thus affecting the architectural platform. Other values that were pursued were for example a sellable product which is easy to place, pleasant living spaces, and a sense of quality, which are values affecting the architectural and technical platforms. Fast project specific detailing processes and knowledge accumulation in the organisation was also considered valuable. At the first meeting the development group discussed and decided on the main characteristics of the house product. This included that the building should be a tower-block with a space efficient stairwell and with space efficient apartments, since this would make the building easy to place and lead to cost efficiency for the end user. A space efficient stairwell led to the decision of maximum eight stories, since the Swedish building regulations limits the building height when using open stairwell solutions. The development group also decided on what building system to use, and that the installations should be centred as close to the stairwell as possible, aiming at an efficient production process. All solutions that were selected had been tested in previous projects and assessed to fulfil the value demands of the architectural, technical and process platform. The plan was never to develop innovative ideas, but instead to use those methods and solutions that were known to deliver value for the client, end user and NCC. Multiple solutions for each functional demand were therefore 6

Paper I

rarely discussed. During the meeting, the discussions were open for everyone in the group to have a say, using a kind of open brainstorming process. When the main concept was decided, the work with the layouts of the apartments started. This was an iterative process where the architect made suggestions, presented them to the development group, got critiques and adjusted the drawings for the next meeting. In order to keep the building easy to place, a maximum framework for the expansion zone was suggested by the architect. The expansion zone made it possible to use configurable apartment modules of different sizes to enable product variety, while still keeping the same building system and technical solutions. The idea of the expansion zone meant an iterative step back to the task specification, changing the content of the building from small sized apartments to configurable apartment modules. This iteration was not time consuming since the task specification was of general character and since it did not affect the decisions already made regarding the main concept. The iterative process of developing the floor plan continued until everyone was satisfied with the result, providing more and more detailed descriptions of the solutions. When the architectural modules, and thus the floor plans were finished, the group started with the detailing of what would become their base model. It was based on one of the layout drawings from the previous stage. No structured methods were used for the development process, but the project leader managed the progression of the work in terms of time aspects. The detailing process was carried out like in ‘traditional’ house-building projects, assigning subcontractors and working out the technical details, resulting in project documents for production. The process platform was not documented in this stage, since the product related processes were well established among the blue and white collar workers in Umeå. It was soon realised however, that this was not sufficient when expanding the sales outside of Umeå, and the process platform was developed as support for project planning and production. The continuous development of the Folkboende house-building platform is handled by the Folkboende Group at NCC in Umeå. They develop the platform based on feedback from projects, their own knowledge of the platform and on customer surveys. In this way the Folkboende Group can collect information about how the buildings are perceived by the customers and thus keep them attractive and up to date.

Advancement in the views of developing house-building platforms Analysing the development process of NCC Folkboende to product development approaches, a framework for describing the development process of a house-building platform with a high level of pre-engineering is presented, Fig. 2. Iteration between architectural, technical and process views through CE 7


before the CODP was a new development strategy for the group. It gave, according to the group members, a method of capturing experiences from different value perspectives over and over again, not just once to secure a specific building project. Separating the development process of the housebuilding platform from that of individual house projects, which is necessary according to Jansson et al. (2014), enabled a product oriented development process based on experiences and forecast. By analysing architectural values before CODP to develop platform predefinitions for layout modules, it was possible to plan production in detail. Even though the development group placed high priority on short lead times and cost efficiency, just like in the development projects studied by Jansson & Lundkvist (2014) and Bonev et al. (2014), the integration of building architecture also ensured tracking of customer and end user values throughout the process. This was considered a prerequisite by the development group when designing this kind of pre-engineered house product. On the other hand, leaving the architectural iteration after decoupling point limits production benefits but could open up for layouts customised for the site and higher flexibility towards customer demands. The CE approach meant that all design solutions were analysed from architectural, technical and process perspectives, which minimized late stage design changes and ad hoc solutions in individual projects. The study showed how structured design approaches, as for example the one described by Pahl et al. (2007), together with an in depth understanding of design iteration, might be a first step towards a more controlled development process of house-building platforms. The development stages for NCC Folkboende were taken in sequence but with a repetition by iteration (arrows in Figure 2) in order to reach specified values embedded in the different views of the platform. Even though the development process of NCC Folkboende has been shown sequential in design activities, the activity outputs were seldom used other than intuitively. Neither was the progression verified against initial requirement specifications and value objectives in a structured way. The requirement specification in combination with activity results could have been used throughout the process to ensure that design solutions are optimised (Cross, 2008), and in this way limiting the risk of excluding important value perspectives. The fact that the platform-organisation has identified the importance of continuous development through experience feedback is a key factor for the long term success of the house-building platform. In this way the changing market demands can be met and new technology can be taken into account by incremental steps instead of reengineering new products in order to improve the concept and its related processes (Jansson et al., 2014).

8

Paper I

Figure 2. Structure of the development process of NCC Folkboende, analysed with the general design approach described by Pahl et al. (2007).

Conclusions By analysing the development process of NCC Folkboende against the general design approach described by Pahl et al. (2007), it has been possible to show how house-building platforms with a high level of pre-engineering can be developed in a sequential manner and in accordance with product development theories. The research also show how iterations between architectural, technical and process solutions, through the use of CE, enabled product definitions before the CODP that captures values specified for NCC, property owner and end user. The conclusions are further elaborated by answering the research questions in the following sections.

9


How structures for product development could contribute in the development of house-building platforms It has shown possible to compare the structure of the development process of a house-building platform to the general design approach described by Pahl et al. (2007). Since structured methods are positive for product development projects (Graner & Missler-Behr, 2013) this kind of organised development process can be used to ensure the progression of a development project. It also decreases the amount of which the development process is based on intuitive methods. By applying product development methods, advancement towards time-efficient development processes that tracks quality, user demands and company profit can be achieved. By defining requirements, design activities, results in form of activity outputs, and methods for measuring requirement fulfilment, the development process could be communicated to the development group in a way that clarifies goals and supports teamwork. A defined development structure provides a basis for capturing value perspectives throughout the process, and for routinizing design processes in future platform development processes. How values that are embedded in a platform contribute to the development process The complexity of platform systematisation, as described by Voordijk et al. (2006), was handled though the use of CE. By including the building architecture in the pre-engineering of the platform, it was possible to optimize the building definition by balancing architectural and technical values with production efficiency, thus creating economy for both NCC, property owner and end user. The lack of tracking progression against initial requirement specifications and values became totally depended on the iteration between participants instead of a supporting structure. The multiple perspectives of each design solutions also limits the risk of ad hoc solutions during production by continuous improvement, but with a focus on process values.

Acknowledgements The authors would like to acknowledge the contribution from NCC Construction in Umeå, Sweden, who provided the possibility for the data collection. We would specially like to thank those who participated in the interviews and those who in other ways contributed to the work process.

10

Paper I

References Akao, Y., 1990. Quality Function Depolyment – Integrating Customer Requirements into Product Design. Cambridge: Productivity Press. Apleberger, L., Jonsson, R., Åhman, P., 2007. Byggandets industrialisering: Nulägesbeskrivning, Göteborg: Sveriges Byggindustrier. Bonev, M., Wörösch, M., Hvam, L., 2014. Utilizing platforms in industrialized construction: A case study of a precast manufacturer. Construction Innovation, 15(1), pp. 84-106. Brege, S., Stehn, L., Nord, T., 2013. Business models in industrialized building of multistorey houses. Construction Management and Economics. 32(1-2), pp. 208-226. Cross, N., 2008. Engineering Design Methods: Strategies for Product Design. Fourth edition. Chichester: Wiley Evbuomwan, N. F. O., Anumba, C. J., 1997. An integrated framework for concurrent life-cycle design and construction. Advances in Engineering Software, 29(7-9), pp. 587-597. Gibb, A. G. F., 2001. Standardization and pre-assembly: distinguishing myth from reality using case study research. Construction Management and Economics, 19 (3), pp. 307-315. Gosling, J., Naim, M. M., 2009. Engineer-to-order supply chain management: A literature review and research agenda. International Journal of Production Economics, 122(2), pp. 741-754. Graner, M., Misler-Behr, M., 2013. Key determinants of the successful adoption of new product development methods. European Journal of Innovation Management. 16(3), pp. 301-316. Jansson, G., Lundkvist, R., 2014. The role of experience feedback channels in the continuous development of house-building platforms. Construction Innovation, accepted for publication 2014-12-23. Jansson, G., Johnsson, H., Engström, D., 2014. Platform use in Systems Building. Construction Management and Economics. 32(1-2), pp. 70-82. Jensen, P., Olofsson, T., Johnsson, H., 2012. Configuration through the parameterization of building components. Automation in Construction. 23, pp. 1-8. Johannesson, H., Persson, J.-G., Pettersson, D., 2013. Produktutveckling: effektiva metoder för konstruktion och design. Second edition. Stockholm: Liber Johnsson, H., 2013. Production strategies for pre-engineering in housebuilding: exploring product development platforms. Construction Management and Economics. 31(9), pp. 941-958.

11


Khalfan, M., Raja, N., 2012. Improving Construction Process through Integration and Concurrent Engineering. Australian Journal of Construction Economics and Building. 5(1), pp. 58-66 Lessing, J., 2006. Industrialised House-Building: Concept and processes, Licensiate thesis. Lund: Lund University. Merriam, S. B., 1994. Fallstudien som forskningsmetod. Lund: Studentlitteratur. Olhager, J., 2003. Strategic positioning of the order penetration point. International Journal of Production Economics, 85(3), pp. 319-329. Pahl, G., Beitz, W., Feldhusen, J., Grote, K.-H., 2007. Engineering Design: A Systematic Approach. Third edition. London: Springer Robertson, D., Ulrich, K., 1998. Planning for product platforms. Slaon Management Review, 39(4), pp. 19-31. Suh, N. P., 2001. Axiomatic design: advances and applications. Oxford: Oxford University Press. Thuesen, C., Hvam, L., 2011. Efficient on-site construction: Learning points from a German platform for housing. Construction Innovation, 11(3), pp. 338-355. VDI-Richtlinie 2221, 1993. Methodik zum Entwickeln und Konstruktion technischer Systeme und Produkte. Düsseldorf: VDI-Verlag Voordijk, H., Meijboom, B., de Haan, J., 2006. Modularity in supply chains: a multiple case study in the construction industry. International Journal of Operations & Production Management. 26(6), pp. 600-618. Wikberg, F., Ekholm, A., 2011. Design Configuration with Architectural Objects. Ljubljana, eCAADe Conference Proceedings, pp. 451-460. Williams, C. B., Allen, J. K., Rosen, D. W., Mistree, F., 2007. Designing platforms for Customizable Products and Processes in Markets on Non- Uniform Demand. Concurrent Engineering Research and Applications, 15(2), pp. 201-2016. Yin, R.K., 2014. Case Study Research: Design and Methods. London: SAGE.

12

Paper II

Paper II

DESIGN MANAGEMENT USING KNOWLEDGE INNOVATION AND VISUAL PLANNING Author’s manuscript Gustav Jansson, Emma Viklund and Helena Lidelöw Published in Automation in Construction 72 (2016), pp. 330–337 http://dx.doi.org/10.1016/j.autcon.2016.08.040

Abstract An open platform used for industrialised house-building imposes restrictions on the flexibility of the product offering when developing design standardisation. How design process standardisation incorporates variations in products has not been widely studied. The aim of this research is to explain how design breakdown enables Lean Product Development Flow (LPDF) and lookahead planning in an industrialised house-building context where an open platform is used. A case study was conducted of how one of the leading industrialised housebuilding companies in Sweden introduced the LPDF tool Knowledge Innovation/Visual Planning (KI-VP) into their design process. The implementation of KI-VP led to an increased cross-functional understanding of relationships between activities, which are an important factor in achieving flexibility and a synchronised workflow. By using design standardisation, lookahead planning was implemented and used in the management of design flow. Standardisation through design breakdown provides a basis for knowledge innovation that enables improvement of the open platform using a bottom-up approach and increases the production flow.

Introduction Providing a turn-key solution for a client forces a contractor to optimise the product, from design and manufacturing through to assembly and delivery of the finished building. When the contractor controls the design, there are many opportunities for them to improve the design process using work breakdown structures and lean principles [1]. For residential construction, the use of either an open or closed platform makes it possible for the contractor to produce a standardised work process [2]. The platform is a collection of the contractor's assets, used repeatedly for different construction projects, thus creating industrialised processes [3]. When closed platforms are used, the design process 1

Design management using knowledge innovation and visual planning is part of production and is organised using configuration tools based on modularisation [4]. Using an open platform, where the client can specify requirements outside the configurable solution space, is a challenge since such variation affects design and production planning [2]. Tribelsky and Sachs [5] identified how stable design information flow could be managed to handle variations in construction design, through mechanisms such as creating small batch sizes, maintaining small quantities of work in progress, by having short design review and response cycle times, as well as identifying and removing bottlenecks. There has been little research about the breakdown of design processes into activities that could be used to manage the information flow during the design phase in industrialised house-building. Lean Product Development theories and methods, described in manufacturing literature, have been used in this research as a lens to capture design process standardisation. For example, the use of visual planning of standardised work tasks, in combination with experience feedback, is known to enhance the transparency of processes. Continuous improvements in design can be achieved by combining various lean methods to create a visualisation of knowledge creation [6]. This technique forms the practical method called “Knowledge Innovation/Visual Planning” (KI-VP) [7]. The lean method of look-ahead planning provides another practical approach, focusing on achieving precision in the planning of the design process [8]. A combination of practical Lean Product Development Flow (LPDF) tools and look-ahead planning methods form the analysis of the design management of house-building platforms. The aim of this research is to explain how the management of design breakdown enables the use of LPDF in combination with look-ahead planning in an industrialised house-building context where an open platform is used. A case study investigated how one of the leading industrialised house-building companies in Sweden implement and use KI-VP in their design process.

Frame of reference Lean Manufacturing strategies have evolved from the Japanese manufacturing industry and are widely used to improve manufacturing, with the primary aim of identifying and eliminating waste across the entire value chain [9]. The need for efficiency in the product realisation chain places demands on actors to understand the requirements of their customer's customer and the conditions for their suppliers' supplier [10]. In industries producing complex and expensive products (such as house-building), product development is often part of the design process [11]. LPDF is described by Oppenheim [12] as being based on the five principles of lean for design management: value measured using throughput time for the 2

Paper II

customer; the use of takt time by mapping the process in an Obeya room to give current and future states; regular face-to-face meetings to minimise waste in the flow; identification of internal customers and understandable deliverables to enable pull in the value stream, and team training to continuously improve planning and product quality in the pursuit of perfection. An Obeya room is a room where ongoing activities are visualised and design meetings held. By manufacturing small batches of products, with frequent transactions and fewer interruptions, stable design information flows and a higher quality of construction design documents can be achieved [5]. As a practical application of LPDF, the concept of KI-VP was incorporated by Japan Management Association Consultants [7]. The KI-VP concept is based on having an Obeya room, planning schedules on boards, the use of standard operations sheets (SOSs) capturing activities, daily experience feedback, and process analysis. When implementing the KI-VP concept with tools, techniques and methodologies, Hines et al.[13] discussed the need for understanding both lean thinking at the strategic level and lean production at the operational level, to create knowledge flow within and between projects. They described the approach as bottom-up, where the participants in the process are the people who establish the existing process state, decide what to improve and how to improve it. To handle tight development schedules and support queue management, Rossi et al. [14] suggested the idea of visual planning in an Obeya room. Visual planning is a method that allows the simultaneous planning and visual communication of streamlined work processes and the interdependencies between activities [9].Themethod is simple: activities and deliverables are outlined and illustrated on a physical planning board (often a whiteboard) and updated at frequent meetings in the Obeya room [15]. Visualisation and physical movement of design activities on the board give a sense of flow, while the frequent meetings provide an environment to share knowledge and discuss problems [15]. The use of magnetic visual boards is often the first step in the implementation of a lean strategy for service processes to identify bottlenecks, obtain operational transparency, and enable a fast visualisation of flow-related problems [16]. Visual boards are used to display the process, the progress of projects, and opportunities for improvement [17]. In a case study carried out by Viana et al. [6], it was shown that visual planning can reduce the complexity of process management in an industrialised construction context, where multiple projects are designed simultaneously. An SOS is a detailed listing of all work tasks in an activity related to a particular stage. It is a lean manufacturing tool that aims to improve the process as a whole by standardising parts [18]. By capturing the current lowest level with an SOS, the activity can be improved, leading to the creation of a new SOS. 3

Design management using knowledge innovation and visual planning

In a tool-based analysis of visual planning in construction, Viana et al. [6] identified an increased adherence to planning targets when companies worked with work breakdown structures and visual boards. The idea behind using planning boards is the visualisation of the company's goals for all production units. Little is known about the process of capturing production knowledge and developing it as a source for improvement of design work in house-building companies [2]. When managing projects, the long-term investments in integrating design and construction competences are traded against short-term efficiency and long-term innovation [19]. Short-term efficiency focuses on project performance; distributed work practices limit incremental innovation from project experience [20]. Long-term innovation can be achieved through continuously developing a house-building platform to create robust design work with feedback channels [2]. “Knowledge Innovation” refers to using the upstream flow of experience feedback and the transparency of visual planning to learn and improve the process [13]. Hines et al. [13] described how each activity can be analysed to help fulfil project goals, and also why unscheduled activities occur and whether they should be included in KI-VP standardisation or not. The resolution of activities described in the SOSs is difficult to visualise in an overview on a physical planning board since there can be hundreds of activities in a process. The use of manual boards and physical artefacts e.g. magnets, presents an overview but can make it difficult to store and share knowledge [15]. Daily meetings and the breakdown of activities contributes to the enhancement of operational integration, cross-functional communication and visualisation of potential problems [17,21]. Using a design breakdown structure, the decomposition of processes into activities organised by size, duration and responsibility can be detailed in task descriptions covering the questions what, who, when and how [22]. In this way, it is possible to define a master schedule template that can be used as a basis for planning and process improvement [13]. The development of look-ahead planning in the field of lean construction [23] resulted in a method of improving the project performance and creating a predictable flow in production that was realisable. Look-ahead planning consists of a master schedule, a look-ahead schedule and a weekly work plan [23]. The master schedule is defined at the beginning of each project, based on status and forecast information, enabling an overall view of the project in the look-ahead schedule. The plan for a few weeks ahead is refined, enabling coordination between design activities and deliveries. Activities are further refined in the weekly work plan, which usually specifies what these activities are, but not how and why to carry them out [24].A case study of a hospital project Hamzeh et al. [8] showed how look-ahead planning was applied to plan the information flow in the design process. The complexity of identifying the design process is due to 4

Paper II

the interdependency of activities, numerous design iterations, and the need to continually re-plan [8].

Methodology A case study approach was used since it gives an in-depth view of the studied phenomenon [25]. Data were collected by combining interviews with data from documents and observations. A single case provides the opportunity to make indepth analysis of relationships and meanings in the context at the loss of generalisability. Therefore, case study results are compared to existing theory when drawing conclusions. The most successful industrialised contractor in Sweden was chosen as the company to be studied due to their extensive use of lean thinking in design and production. The case study was longitudinal, with data about the structure of visual planning and its use in daily work being collected between 2006 and 2010. Building on that study, the company was revisited in 2015 to discover how their KI-VP implementation had developed since 2010. The first data collection in 2010 involved the mapping of the design breakdown based on personal time reports, external consultant invoices, site visits and project-specific time reports for 26 building projects between January 2005 and June 2009. The implications of design activities, and the relationships between them, were identified through observations and notes from project meetings in the Obeya room at the company during the process of refining the standardised design work. By tracking how much time the employees spent on each activity and the interdependence between activities, it was possible to breakdown and analyse the design process for standardisation. To analyse the flow, the starting point for the process analysis was the first manufacturing step in the factory and from there each delivery and activity was traced backwards through the design process. In 2015, the improvement strategy and the implementation of the KI-VP system were recorded using site visits and interviews with two project managers and the lean coordinator at the company. The sales manager was also interviewed regarding how the sales process was organised from a flow perspective. By using a semi-structured approach in the interviews, it was possible to capture the interviewees' thoughts on the design breakdown process, and to ask follow-up questions [25]. All interviews were recorded and fully transcribed. The content of the transcription was then categorised, based on the timeline of the development and refinement of the KI-VP system. The predictability of design planning was discussed using data gathered between 2006 and 2012 from 62 projects, recording design time divided by square meters of area built. Further analysis was made of the LPDF strategies and the 5


description of look-ahead planning methods, focusing on how their practical application affected the design work flow. To explain how design breakdown functions using a combination of lean construction and lean manufacturing methods, the practical use of KI-VP was analysed using Liker's [9] perspective of standardisation, implementation and flow. The unit of analysis here was the design process at the case company.

Case company The case company produces wood-framed residential buildings including student apartments, rental apartments, condominiums, nursing homes, and residential accommodation for the elderly. The buildings are factory-produced in volumetric modules, which are assembled on the construction site and are manufactured within the company, see Fig. 1. An open platform containing technical solutions for elements such as façades, building frames and modular interfaces, is used in all projects to enable a combination of repetitive and unique design solutions. The sales department use design-build contracts only. This turn-key approach has led the company to strive towards optimisation of the entire process in order to reach an efficient flow of projects through the production process. The information flow in the design process is based on downstream pull and was identified as a bottleneck that limits the takt time of production. This motivated the introduction of lean thinking into the design process. The design includes systems design, detailed design, and production planning. Conceptual design is carried out by external architects. Part of the conceptual designs are based on the limitations of the case company's production system and some of them have had to be adapted retrospectively to fit module sizes, span lengths, wall thicknesses etc. How well the platform is integrated into the conceptual design depends on when the case company enters the conceptual design process, and also the architect's familiarity with the production system. The collaboration with external architects means that some solutions are developed before the case company starts their design. In this way, the architect can develop solutions based on a project vision. Important architectural values such as aesthetics, functional relationships, integration with the surrounding environment etc. are defined prior to the engineers' design work with the housebuilding platform and the technical details. Architects' are finalised at the first design meeting at the case company. However, it has become increasingly common that there is occasional collaboration between architects and the case company engineers during design, especially when issues arise regarding platform constraints that has not been incorporated into the conceptual design.

6

Paper II

Figure 1. The product realisation chain at the case study company.

KI-VP implementation The case company started to work with visual planning and design breakdown in 2002. The initial state was a structure, based on a rough time plan where the production start in the factory was the deadline for each project. Even though some milestones for time management had been identified, for example the delivery time for windows and stairs, the multiplicity of projects made it difficult to capture and optimise the design process. As a first step towards ensuring better control of the progress, the project managers developed a checklist where they used colour codes to identify whether the main deliveries had started or not. Even though the list only provided a rough description of deliveries, it served as a means to ensure none were overlooked. In 2007, they made a visual representation of the list for everyone in the design process to see. This was achieved by installing a magnetic board, on which the projects were displayed in rows and the deliveries in columns. Green, yellow and red magnets were used to represent status updates in accordance with an improved version of the previous checklist. By having weekly meetings by the magnetic board, where everyone could explain why their deliveries had or had not been finished according to plan, the awareness of the relationships between deliveries increased in the group. The focus was on whether the deliveries were finished or not, not necessarily on the reasons and effects of delays. To better understand the design process as a flow, the next step was taken in 2010 to map the flow of information and activities using the KI-VP concept, as described in lean Product Development Flow literature. Also, activities that were not pure design activities, for example purchasing and some activities carried out on the construction site, were included in the design breakdown since they were part of the information flow in the design process. Strategic purchasing had a strong influence on time management, and it formed key milestones while structuring the design process. By adding the dimension of time to the visual content on the magnetic board, it was possible to start working towards higher efficiency to match the takt time of the factory. Design break-down resulted in a standardised design process template that worked as the basis for developing the current KI-VP concept. The initial intent was to visualise this master schedule template with the activities that were common to all projects. By allowing the employees to refine the master schedule template while using it, it was possible to have many perspectives and inputs to the system. This eventually led to the 7

Design management using knowledge innovation and visual planning refined master schedule template that is used today (2015), where 400 activities have been specified and scheduled in a 28 week time plan. For each project, the master schedule is visualised on a visual planning board in an Obeya room. Each activity in the master schedule template contains a number of tasks which are, for most activities, specified by SOSs to ensure that these activities are carried out in the same way, regardless of who does them, see Fig. 2. The aimwas to ensure that anyone could perform each work task, regardless if they were used to performing the task or not. The SOSs have been developed by the employees themselves since the start of 2014, in order to ensure that everyone agrees on the content and time of each delivery. The clearer the agreements of deliveries have become, the tighter the time plan has become. The master schedule, together with the visual planning boards, provides a process view of the information flow while SOSs provide detailed descriptions of how and to whom the information should be delivered. The collection of SOSs forms the process asset in the platform, while the technical solutions reapplied between projects form the technology asset. Standard Operation Sheets and the master schedule template are under constant improvement through knowledge innovation, led by the team leaders. The team identifies parts of the process that need improvement e.g. the design of electrical shafts. A subset of the team is assigned the improvement work (which is carried out alongside daily operations) and the results are documented in the platform and/or in an improved SOS. In order to maintain the standardised design process, team leaders observe how the SOSs are used in practice. This method was implemented at the start of 2015 with the expectation that it would give continuity to the process and enable examination of why changes from the SOS occur. The 28 week master schedule template is used as a base for all design processes. It is adapted into an individual master schedule for each project, since the design phase rarely takes the full 28 weeks and the projects vary in

Figure 2. Extract from the case company's master schedule template, showing a schematic view of activities with different disciplines in different colours, an example of activities, and how each activity is defined by standard operations sheets. 8

Paper II

complexity and size. This adaptation is carried out manually by the project managers, by adjusting the time span and regrouping the activities. A risk, that has been identified, is that of overlooking links between activities which might result in delays in the process or missed sub-deliveries. Therefore, it has been suggested that better visualisation of the links between activities or automation of the revision of the master schedule template may improve matters. The master schedule template is revised on an annual basis, synchronising new activities and deliveries. The visual planning boards have magnetic markers for each activity, which is manually put in place in the Obeya room, see Fig. 3.This is recognised as being time-consuming for the project managers, since they have to clean the markers and replace them before the start of each new project. The staffs want the company to invest in a digital system to handle the visual planning boards. However, before this investment is made, the management want to refine further the implementation of the work method. Up to seven projects are handled simultaneously in the design process. The design team (18 people in 2015), together with representatives from the purchasing and factory departments, gather around the visual planning boards in the Obeya room each morning to carry out a daily review of all projects. The meeting is short: each project leader reports problems and changes in deadlines for their projects. Detailed discussions on solutions are held outside the meeting. Revised drawings are issued to the purchasing and production departments. Changes in bills of materials are reported to the purchasing department. Experience from factory production, on-site assembly, and purchasing is fed back using an established system that involves the entire company. This has led to a better understanding of the design process and the team has seen their work become more homogenous. Process analysis, based on experience feedback, accomplishes monthly because the KI-VP has enabled a more structured work approach. Instead of “firefighting”, the design team now adopts a preventive approach. By physically updating them each morning, the visual planning boards

Figure 3. Obeya room at the case company, using one magnetic board to accommodate each project. 9


have become live documents and problems surface easily, as is the objective of the KI-VP method. Since each activity and delivery is specified in the SOS and in the master schedule template, it is possible to trace delays and faults back to their root cause. This has served as an incentive to reduce the number of delays, since the person who created the delay will be the one to take responsibility for it. The KI-VP concept has, however, turned out to serve as a memory-list and a support system to question the normal state instead of developing continuous process improvements and finding the root causes of mistakes. This was expressed as a limitation of current work methods and one of the improvement areas for the future. One of the interviewees explained that the design team has become well trained in the visual planning part regarding planning projects and following up on progress, but still has some way to go regarding knowledge innovation i.e. experience feedback and process analysis. In order to further improve the KI part, the interviewee explained that it is important that the management take the lead. Even though there are still some parts that can be improved in respect of the use of KI-VP at the case company, positive results for planning predictability have been achieved. Time measurements of the first 21 projects, made between 2006 and 2009, showed a mean value of 0.70 h/m2 spent in the design process with a standard deviation of 0.36, compared to the 28 projects in 2010–2013, where time decreased to 0.62 h/m2 with a standard deviation of 0.23. Of the 63 projects studied in total, 7 were outliers with 1.24–1.72 h/m2 that had limited reuse of technical solutions, unspecified client requirements before design, and architects that were unfamiliar with the platform standardisation. The case company increased the resolution in the visualisation of the design process, from 17 main activities to 400, over a period where the turnover increased from 14 million Euros in 2006 to 75 million Euros in 2015. Takt time in factory production has increased from 30 to 32 modules per week in 2007 to 50–52 modules per week in 2016. According to the interviews with project managers, the hours/m2 spent in design are almost the same currently but there are fewer outliers and predictability is higher. Both the lean coordinator and one of the project managers confirmed that there is the potential to improve the KI-VP by tracking and developing the design process even further. The number and variety of projects undertaken were important for the definition of activities and work tasks in the standard operation sheets, both to obtain enough resolution and to create flow in the process. The sales process is a creative process where clients hire architects to define and present the vision of a project. If the architect knows the predefined standardisations in the open platform or is involved early in the sales process, fewer open parameters are left to resolve in the design processes. According to sales representatives, the matchmaking between the production line and 10

Paper II

potential building projects is a question of how to visualise the flow before design.

Analysis Visual planning boards provided the first step in implementation of LPDF strategies. As Olausson and Berggren [16] described, the boards produced the anticipated results, such as identification of bottlenecks, operational transparency, and fast identification of flow-related problems. In turn, the standardisation of the design process, in terms of the master schedule template and the SOSs, provided the foundation for knowledge innovation, where the current state could be continuously questioned and refined [13]. At the case company, this was achieved through daily experience feedback that involved the entire company, and through process analysis done outside of the actual design process, see Fig. 4. Since the implementation of KI-VP, the design team has experienced a more homogenous process where the complexity described by Hamez et al. [8] is managed. This agrees with Tribelsky and Sacks [5], who suggested that a stable stream of information in projects throughout the design process could result in a synchronised work flow and also that unstable information flows in the design phase have different velocities that produce unpredictable project outcomes, see Fig. 4. Up to seven projects can be handled simultaneously in the design process, each having different time spans due to the variations made possible by the open platform. This means that the standardised design activities are sometimes rearranged and/or compressed/stretched in order

Figure 4. Analysis of the KI-VP implementation at the case company. 11


to accommodate different velocities through the process. The synchronised work flow continues from design, through production, to assembly and completion of the finished building. This provides the basis to achieve the difficult task of a structured experience feedback throughout the value chain. The difficulties in storing and using experience feedback challenge an organisation to improve the upstream flow of knowledge [20]. KI-VP provides channels for this kind of flow through daily experience feedback to constantly improve the master schedule template and the open platform which, according to Hines et al. [13], is the purpose of knowledge innovation. Even though the design team admit to not having fully implemented knowledge innovation yet, it is clear that KI-VP has helped increase the cross-functional knowledge in the company. The focus has, however, been on information flow and waste reduction, leaving customer values to be considered at the sales and conceptual design stage. LPDF builds on visualising the flow, making it transparent so that the participants can understand and improve it [6,12]. At the case company, lookahead planning is incorporated in the KI-VP tool. The defined master schedule template is the basis for the master schedule for each project. Instead of having to specify activities in look-ahead schedules and weekly work plans for each project, all activities are built into the 28 week master schedule template, thus having the detailed work descriptions automatically linked to all schedule hierarchies. Using SOSs, the defined work tasks become even more detailed than in regular weekly work plans, and thus provide a solid support system. In planning, the visual boards are used as live documents enabling look-ahead planning. The standardisation of design activities in the open platform simplifies the planning process, since the activities only need to be rearranged and/or compressed/stretched depending on the project-specific circumstances. Both look-ahead planning and LPDF supports the values that Lindlöf and Söderberg [15] described about using a physical room for the daily meetings to ensure design progress. The participants in the daily meetings at the case company were not only the design team, but also their closest internal customers, the purchasing and the factory departments. This is a cornerstone for fast experience feedback of work performance. The Obeya room is also a place to critically analyse the design work against expected goals. The case company has shown that it is possible to implement LPDF in an industrialised house-building context where open platforms are used, see Table 1. The study also shows that it is possible to standardise large parts of the design process, when this process includes systems design, detailing and planning for production. In this way, look-ahead planning can be incorporated and facilitated in the open platform. This has enabled a lean workflow through design, synchronised with the takt time of the factory production. The predefinition of building components and related processes 12

Paper II

provides a basis for knowledge innovation that enables a long-term perspective of the improvement of an open house-building platform. Table 1. Lean principles in LPDF and the implementation of KI-VP at the case company. Lean principles in LPDF [12]

Case company's implementation

Value measured by throughput time for the customer

The implementation of KI-VP at the case company has led to a structured design process where the design team works in collaboration across functions to define and improve activities. This has led to clear agreements on responsibilities, on what should be delivered, and when and how these deliveries should be made. The clearer the agreements become, the less throughput time is necessary in the design process and the work process becomes more homogenous.

The use of takt time by mapping the process in an Obeya room for the current and future states

Introducing KI-VP was a way of matching the design process to the takt time of the factory and, in this way, eliminating the design bottleneck. By providing tools to allow better understanding, planning, and follow-up of the process, this elimination has been achieved. The master schedule template provides a basis for structuring each design process. Even though the full 28 week master schedule template is seldom used as is, it provides a structure that can be rearranged to fit the necessary time span and that enables a synchronised working process. The use of an Obeya room makes everyone in, and related to, the design process involved in the follow-up of the projects.

Regular face-to-face meetings to minimise waste in the flow

The Obeya room provides a multi-functional environment meant to increase the awareness of all process steps. By keeping the daily meetings in the Obeya room short and structured, important issues can easily surface without having detailed discussions of possible solutions. Those discussions are held outside the meeting instead and only the affected parties are involved. Thus, the time of unaffected parties is not wasted. An increased awareness of the interrelations between activities, in combination with structured meetings and work processes, minimises waste in the work flow. 13


Identify internal customers and understand deliverables to enable pull in the value stream

By analysing the pull point for deliveries, it was possible to identify the value stream through the design process, starting at factory production and tracing the deliveries backwards through design. In this way, the design process was broken down and each design activity could be specified using SOSs. These include a specification of who will carry out the activity, how the activity should be carried out, who will receive the delivery, what the delivery should contain and when it should be delivered.

Team training to continuously improve planning and product quality in the pursuit of perfection

As expressed by the interviewees at the case company, the design team has become well trained in the implementation of visual planning. They are also working towards better implementation of knowledge innovation, but are yet to reach its full potential. Their work with KI-VP has led to a cross-functional understanding of the design process. This has led to fewer mistakes in the process and the design team has found it easier to ensure the quality of the finished product.

In the case study, two actions interacted to create the KI-VP concept: knowledge innovations carried by daily experience feedback and process analysis (bottom-up implementation) and the visual planning strategies from the lean coordinator (top-down implementation). Together, they created the basis for information and knowledge flow for a continuously developed open platform, see Fig. 5. Implementation of the KI-VP concept at the case company was a combination of strategic investment and operational engagement to improve routines in daily work. The human-centred approach, identified by Hines and Packham [17], with visual boards and daily meetings is used at the case company, supporting a bottom-up process for short-term work. The monthly analysis of experience feedback practiced by the design team creates a bottom-up involvement in long-term innovations.

14

Paper II

Figure 5. KI-VP with top-down and bottom-up movements. Problems are identified through working with visual planning, analysed and solved using knowledge innovation, and then reintroduced to the visual planning. The look-ahead planning approach with different levels of design breakdown is realised by the forecast of activities on the visual board. As described by Hamzeh et al. [8], the design process has to be broken down and visualised for different milestones and deliveries using reverse time scheduling. The uniqueness of the studied case company is that both the design team and the production operators define design deliveries using SOSs in a customer-supplier relationship. Monthly process analysis, carried out by the design team from a bottom-up perspective, makes it possible to improve the handling of daily variations in an open platform. The SOS is the basis of what tasks should be undertaken, who should do them, and when and how they should be done; it also describes the current process state to consider and redefine tasks for. Daily experience feedback flow and monthly process analysis are based on the consideration of what, who, when and how, giving an understanding of the customer's customer and supplier's supplier in a flow-based approach.

Discussion By identifying internal customers from production and upstream, the design breakdown gives a method for standardising a design process for an open platform. The case company started to define their production processes in the factory and thereafter moved on to design by using the KI-VP approach. The identification of internal customers was not the focus when the first visual planning board was set up to visualise milestones. With a higher resolution of design activities, identification of deliveries becomes important and the value of identifying internal customers grows. When engineers, operators and construction labourers identify their own SOSs, the importance of identifying 15


their customers becomes obvious by asking the questions what, who, when and how. The work with KI-VP at the case company included the product realisation chain from design and downstream. A cross-functional focus was necessary, since the entire product realisation chain in the company is controlled by takt time. It is therefore important that all functions integrate and work in harmony with each other. However, this integration is yet to include the Sales and Conceptual Design functions. A problem is that working in Sales means working with projects that do not have a customer order i.e. there are doubts as to whether the project will actually start or not. Thus, it is difficult to plan a pipeline of information as with the design process. The integration of Sales and Conceptual Design in the lean processes at the case company demands different tools to those studied here. The variation of time spent in design arises from a number of input parameters. The size and floor plan of the project affects the use of repetition in design but parameters such as the architect's knowledge of the platform and its technical solutions have a significant impact. Using an open platform creates variation in projects and the management of planning and control of activities. Variation and standardisation have to be balanced in the design process. The case study shows how predefined activities and tasks can be planned and continuously updated through design because of the flexible master schedule template. Even if the master schedule template is set to last 28 weeks, the input knowledge of the incoming project gives a more accurate time plan, taking into account project size, architectural involvement, client requirements, repetition in solutions and building regulations. To match these variations, the platform needs to have flexibility of technical solutions. The variance of technical solutions affects the design work, from organising the master schedule to designing the detailed solutions. As a basis for standardisation, daily experience feedback creates a standardised method for updating the design, purchasing and production phases in a process platform used for industrialised house-building. A cross-functional focus was not easy to create but the KI-VP operations engaged several disciplines to gather knowledge in the SOSs using daily feedback. The SOSs provide a very clear basis for industrialisation. Once stable, the tasks they describe can be standardised and even automated. Care must be taken not to attempt standardisation of work tasks that need to be varied in the open platform. KI-VP has, at the case company, provided a tool for systematic improvements both at strategic and operational levels. If standardisation is introduced in a solely top-down way, there is a risk that the new methods or processes will be perceived as being alien by the employees. The combination of bottom-up and top-down approaches has created an environment where 16

Paper II

everyone who is in, or affected by, the design process is engaged in the standardisation effort. At the company, the refinement of the design process and work methods involves everyone working in the process, but also other functions, for example production, that have deliveries to, or receive deliveries from, the design process. A knowledge flow throughout the company has been achieved. Since this study only covers one single case, it is hard to generalise the implementation process in other circumstances. It is, however, possible to explain and provide an in-depth understanding of how this particular company has worked towards design standardisation through the implementation of lean principles. Managing design work in construction involves people with knowledge and managing this so that the balance between new technical and sustainable solutions for end-user values overlap with proven product solutions and efficient design processes. The study of LPDF at the case study company explains how visualisation with knowledge feedback of design work could be continuously developed to give a better flow for the product realisation.

Conclusions It has been shown that it is possible to apply LPDF, through the use of KIVP, to industrialise house-building, improving the production flow by managing the design process to follow production demands. This implies that lean product development theories can be adapted to fit an industrialised house-building context where open platforms are used, even though the construction and manufacturing fields are governed by different prerequisites. This idea has been further validated by the results of the case study carried out by Viana et al. [6] at a steel fabricator for warehouses and industrial buildings. An open platform enables great variation of the final product and this puts demands on flexibility in the design planning while working towards design standardisation of the process. KI-VP leads to an increased cross-functional understanding of relationships between activities, which is an important factor for accommodating flexibility. Since the 28 week master schedule template at the case company is seldom used as is, the need for a flexible process is obvious. The standardised design activities make it possible to incorporate look-ahead planning into the KI-VP tool. In this way, an overall view of the entire design process is possible, where the definitions range from detailed work task descriptions to overarching project management. This facilitates project planning and look-ahead work methods. The implementation of KI-VP at the studied case company combined topdown and bottom-up approaches, a prerequisite for the usability of the concept. By incorporating a bottom-up approach in the implementation, the defined 17


design process and working methods were accepted by the operators, since the operators were the people to define them. The bottom-up approach also contributed to a clear view of the actual information flow in real projects, and the contents of all necessary design activities. The tools used at the case company to implement lean strategies in the design process focused mainly on reduction of waste and on finding and managing the flow through the process. They did not, however, focus on finding and securing end-user values for the product. These kinds of values are often managed by the architect at the conceptual design phase. A result of this study shows that both conceptual design and sales processes require tools to be developed to manage the flow for their work. What these tools are, and how the architect and sales personnel could be involved in an industrialisation process, is a subject of further research.

Acknowledgements This work was carried out within the research project ATTRACT. We gratefully thank the interviewees at the case study company for their time and inspiration. This work was funded by the Swedish Governmental Agency for Innovation (VINNOVA).

References [1] E.S. Norman, S.A. Brotherton, R.T. Fried, Work Breakdown Structures: The Founda-tion for Project Management Excellence, John Wiley & Sons. ProQuest ebrary, Hobo-ken, NJ, USA, 2011. [2] G. Jansson, H. Johnsson, D. Engström, Platform use in systems building, Constr. Manag. Econ. 32 (1–2) (2014) 70–82. [3] J. Lessing, L. Stehn, A. Ekholm, Industrialised house-building – development and conceptual orientation of the field, Constr. Innov. 15 (3) (2015) 378–399. [4] M. Meyer, A. Lehnerd, Section 3: Platform Strategy. The Power of Product Platforms: Building Value and Cost Leadership, The Free Press, New York, 1997. [5] E. Tribelsky, R. Sacks, An empirical study of information flows in multidisciplinary civil engineering design teams using lean measures, Archit. Eng. Des. Manag. 7 (2)(2011) 85–101. [6] D. Viana, C. Formoso, J. Wesz, P. Tzortzopoulos, The role of visual management in collaborative integrated planning and control for engineer-to-order building systems, Proceedings of the 22nd Conference of the International Group for Lean Construction, Oslo, Norway, 2014.

18

Paper II

[7] T. Tanaka, Efficient Creativity: JIT for Knowledge-Workers, JMAC Consiel SpA, Milan, 2002. [8] F.R. Hamzeh, G. Ballard, I. Tommelein, Is the last planner system applicable to design? A case study, Proceedings for the 17th Annual Conference of the International Group for Lean Construction, Taipei, Taiwan, 2009. [9] J.K. Liker, Section II: the right process will produce the right results, The 14 Principles of Toyota Way: An Executive Summary of Culture Behind TPS, McGraw-Hill, New York, 2003. [10] L. Horvath, Collaboration: the key to value creation in supply chain management, Supply Chain Manag. 6 (5) (2001) 205–207. [11] H. Johnsson, Production strategies for pre-engineering in house-building: exploring product development platforms, Constr. Manag. Econ. 31 (9) (2013) 941– 958. [12] B.W. Oppenheim, Lean product development flow, Syst. Eng. 7 (4) (2004) 352– 376. [13] P. Hines, M. Francis, P. Found, Towards lean product lifecycle management: a frame-work for new product development, J. Manuf. Technol. Manag. 17 (7) (2006) 866–887. [14] M. Rossi, E. Kerga, M. Taisch, S. Terzi, Learning methodologies to diffuse lean product development to industries, IFIP Adv. Inf. Commun. Technol. 388 (2012) 287–298. [15] L. Lindlöf, B. Söderberg, Pros and cons of lean visual planning: experiences from four product development organisations, Int. J. Technol. Intell. Plan. 7 (3) (2011) 269–279. [16] D. Olausson, C. Berggren, Managing uncertain, complex product development in high-tech firms: in search of controlled flexibility, R&D Manag. 40 (4) (2010) 383– 399. [17] P. Hines, J. Packham, Implementing lean new product development, IIE Annual Con-ference Proceedings, 2008. [18] L. Malmsköld, R. Örtengren, L. Svensson, Training virtually virtual, Int. J. Adv. Corp. Learn. (iJAC) 5 (3) (2012) 29–34 Kassel University Press GmbH. [19] P.E. Eriksson, Exploration and exploitation in project-based organizations: develop-ment and diffusion of knowledge at different organizational levels in construction companies, Int. J. Proj. Manag. 31 (3) (2013) 333–341. [20] M. Bresnen, A. Goussevskaia, J. Swan, Embedding new management knowledge in project-based organizations, Organ. Stud. 25 (9) (2004) 1535–1555. [21] I. Gamme, S.H. Aschehoug, Assessing lean's impact on operational integration, Int. J. Qual. Serv. Sci. 6 (2/3) (2014) 112–123. 19


[22] A. Aartsengel, S. Kurtoglu, Chapter 9: create work breakdown structure, Handbook on Continuous Improvement Transformation, Springer, Berlin Heidelberg 2013, pp. 137–138. [23] H.G. Ballard, The Last Planner System of Product Control(Dr.Thesis) University of Birmingham, 2000. [24] H.G. Ballard, Lookahead planning: the missing link in production control, Proceedings for the 5th Annual Conference of the International Group for Lean Construction, Gold Coast, Australia, 1997. [25] S.B. Merriam, Qualitative Research: A Guide to Design and Implementation, Jossey-Bass, San Fransisco, 2014.

20

Paper III

Paper III

MODULARIZATION BASED ON COMMONALITIES IN HOUSE-BUILDING REQUIREMENTS Author’s manuscript Emma Viklund, Marcus Sandberg, Gustav Jansson and Helena Lidelöw Published as proceedings from ICCREM 2016: International conference on construction and real estate management

Abstract Some of the requirements governing the design of houses are common between projects. This opens up for using modularization based on product commonalities. Though modularization is well known in the manufacturing industry, its use in house-building contexts is less studied. Even more scarce is research focusing on how requirement commonalities between one-of-a-kind products are found and managed in a modularization process. In this research, modularization from a requirements management perspective is empirically explored using a case study approach. Though the studied modularization process can be described as sequential, the process steps are highly interrelated, with overlaps and iterations. Commonalities are found by functional decomposition of the customer’s portfolio projects, tracing technical solutions back to their initial requirements. These are balanced with the customer’s requirements on the product and on the design automation tool. The requirement list is continuously updated based on portfolio analysis and communication of modularization results to the customer.

Introduction House-building projects are governed by complex sets of requirements resulting in one-of-a-kind products (Kiviniemi 2005). Jensen et al. (2015) has shown however that there are common denominators in these requirements, if looking at subsystems and components. This makes it possible to do standardizations through modularization based on product commonalities (Jensen et al. 2015). Modularization theories are well-known in the manufacturing industry (Erixon 1998), but its use in house-building is less 1

Modularization based on commonalities in house-building requirements

studied. Two examples though are Malmgren (2010) and Jensen et al. (2015). By defining modules, configuration logics can be used to generate platform based product solutions (Malmgren 2012; Pakkanen et al. 2016). In extension, this enables the use of design automation tools based on configuration logics (Blecker et al. 2004), where design results such as 3D geometry and drawings can be automatically generated. By finding commonalities in an existing product range of a company it is possible to rationalize the product offer while still enabling customization (Pakkanen et al. 2016). In order to achieve such a bottom up approach to modularization, customer requirements on the scope of the module definitions needs to be combined with knowledge already built into the customer’s products (ibid.). Though it is possible to find research on requirement management and modularization, we have yet to find a research perspective where these two are combined in a clear way. A requirement analysis and management perspective in modularization theories would be of use when developing modular products based on generic product structures. Our aim is therefore to explore how customer and design requirements are analyzed and managed in a modularization process for a design automation tool, using house-building as the example.

Background A modular product, in this case a house foundation is used as example, can contain several modules: modules for different subsystems, e.g. edge beams or elevator pits; and variants of a particular module kind, e.g. variants of different types of edge beams (Jensen et al. 2015). These module variants can be parametrical, i.e. possible to stretch or modify to suit specific input conditions without changing the variant’s interfaces (ibid.). A module is built up of one or several components, which according to Simpson et al. (2014), is the most detailed level in the design hierarchy. The process of decomposing a product into suitable module definitions is a modularization process (Erixon 1998). Module based product variants can be generated by e.g. incorporating configuration logics based on a generic product structure (Jørgensen 2001). This allows product optimization while still enabling customization that caters for customers’ needs and values (Jørgensen 2001; Malmgren 2010). According to Zhang (2014) a configurator performs a configuration task and produces a solution together with technical specifications of the design. Elgh’s (2008) description of a design automation system also includes the knowledge behind the module based product design and the configuration logics. Based on these descriptions, a design automation tool (DAT) is here defined as a tool that serves to automate a number of design activities by enabling configuration of module 2

Paper III

combinations, as well as each single module. The tool allows for automatic generation of design results such as geometry representations and drawings, including the knowledge behind the design.

Research methodology In order to explore the topic, a case study approach was chosen (Yin 2014). The case company was selected due to their unique position as developers of DATs for the Swedish construction industry. The aim of the study was to empirically explore how requirements are managed while defining modules based on commonalities in an existing product range. Empirical data was gathered through semi-structured interviews with four of the employees at the case company. These data were combined with demonstrations of the modularization process, observations at the company and document studies of company templates used during DAT developments. The interviewees at the case company work with: engineering and modularization; programming and CAD; modularization, project management and administration; modularization, technical management and sales. An interview guide was prepared prior to the interviews, containing questions such as: “describe the requirements that you manage in your development of the configurator?”, and “describe your process of finding and defining which requirements that affects a configurator?” The questions were formulated to provide an understanding of the case company’s modularization process and development of DATs, and how the engineers work with requirements management. The questions were also designed to cover what kind of requirements that the engineers work with and to explore how the engineers categorizes these. Interesting topics outside the scope of the prepared interview guide which arose during the interviews were also explored in order to ensure sufficient understanding of the modularization process and the development of DATs. All interviews were transcribed and sent to the interviewees for validation. The empirical data was further validated through triangulation (Yin 2014). This was achieved by comparing interview results against each other, against information provided through demonstrations of the work processes, and against documents including checklists and process descriptions. The final description of the empirical data was also distributed to the responsible managers in order to minimize the risk of errors in interpretation. The theoretical framework presented in this paper was used to discuss the empirical findings. The purpose was not to test or validate any theory, but to use the theoretical framework as support for exploring this particular empirical case. The contribution of this paper is therefore focused on the exploration of a practical approach to requirements management in a real case modularization 3


process. This knowledge will be used as one part of our future research on methods for requirement management, modularization and design automation. Case Company The chosen case company is an affiliated company to one of Sweden’s leading construction engineering companies. They develop DATs that enable automatic generation of construction drawings or documents. Automation of design is sold as a service to construction consultancy companies or contractors, thus keeping the DATs in-house at the case company. The products in the DATs are module-based and the design automation is based on configuration logics. The DAT has an end-user interface enabling entry of project specific variable indata. Construction drawings or documents, depending on the particular DAT, are then automatically generated. Even though the case company works with a variety of products, their methods for finding requirement commonalities during the modularization processes are similar, regardless if the end products are e.g. house subsystems, structural elements for bridges, or documents. For the sake of explanation, subsystems for house-building will be in focus throughout this paper.

Theoretical framework Existing methodologies for design automation application development, e.g (Stokes 2001) and (Blecker et al. 2004) mention requirements management. Stokes’ (2001) focus is on capturing and formalizing engineering knowledge, based on the logics of what, why and how. Erixon (1998), describe a modularization methodology (Modular Function Deployment) where requirement management is included. Elgh (2008) developed an integrated system for automated variant design and management of manufacturing requirements. His aim was to capture what Stokes (2001) calls the why and the how, i.e. the rationale of a design process. His view on requirement management includes the process of identifying, formulating, allocating, verifying and managing changes of requirements. In order to enable traceability, IT-support is recommended to be used (Ozkaya and Akin 2007). By incorporating Elgh’s (2008) requirement management view in Erixon’s (1998) modularization process a six step process structure can be derived. The steps are presented in sequence, but according to Erixon frequent iterations occur between them. Step 1, Identifying Requirements: Erixon (1998) explains the necessity of identifying the right customer and design requirements. In order to ensure this,

4

Paper III

Elgh (2008) suggests the use of checklists where stakeholder aspects and life cycle perspectives are included. Step 2, Formulating Requirements: Elgh (2008) focuses on ensuring requirement traceability throughout a modularization process. Erixon (1998) uses Quality Function Deployment (QFD) as a mean to specify and analyze requirements and their interrelations. In this approach, customer and design requirements are separated to analyze the dependencies between them as support when defining the product. In order to ensure meticulous treatment of requirement data, Erixon suggests the use of affinity diagrams to decrease the amount of requirement definitions in the QFD matrix. The approach of grouping requirements in order to make the data set more manageable is also used by Elgh (2008), though using a classification diagram instead of an affinity diagram. Step 3, Allocating Requirements to Technical Solutions: The product is broken down into functions and related technical solutions to find their interrelation with previously defined requirements (Erixon 1998). This functional decomposition generates an understanding of the product and of how each part contributes to the overall function (ibid.). Design Structure Matrices (DSM) and hierarchical Function Means Trees (FMT) are used as decomposition tools to structure the process (ibid.). When analyzing technical solutions during the functional decomposition, all possible technical solutions should be documented in order to evaluate which one is best suited (ibid.). When allocating requirements to technical solutions, traceability of requirements should still be possible. Elgh (2008) therefore systematically gathers all rationale behind a transformation in order to enable subsequent definitions of executable code in the development of the DAT. Step 4, Generating Concepts: The modularization concept is now decided. This is based on the strategy and aim, i.e. module drivers, and includes how the modules should be defined in terms of standardization, parametrical attributes and design variations (Erixon 1998). Variations are kept to a minimum by finding commonalities in sub-functions in the functional decomposition (ibid.). Benefits of standardizing technical solutions based on commonality functions should however be weighed against the potential benefits of variation (ibid.). In order to evaluate what sub-functions to modularize, a Modular Indicator Matrix (similar to QFD but analyzing module drivers against sub-functions) can be used (ibid.). In order to choose which alternative technical solutions to select, Erixon uses a selection matrix to evaluate technical solutions against project goals. The end result of the fourth process step is a few module concepts including a rough estimation of the module design (ibid.). Step 5 Evaluate Concept: At the last two process steps, Erixon (1998) discusses evaluation and module improvements based on a broad perspective e.g. incorporating business goals, economic aspects and aspects such as time, 5


quality, efficiency etc. These aspects are important module drivers, and thus should be considered during the modularization process. Erixon explains that evaluation of the modularized product structure should be done based on structured methods defined in beforehand. Evaluation is performed to ensure that the defined modules enable satisfactory product configuration. Step 6, Improve Modules: In the last process step, Erixon (1998) focuses on final adjustments derived from knowledge developed during the previous process steps. At this stage, manufacturability should also be considered (Elgh 2008), for example by using Design for X approaches (e.g. Design for Manufacturing or Assebly) (Erixon 1998).

Results and analysis Figure 1 shows a graphical representation of the case company’s modularization process based on the theoretical framework of this report. The content of the figure is described in detail below, based on each of the process steps from the theoretical framework. The modularization process is closely linked to the development of the DAT, since requirements on the DAT can have implications on product modularity and vice versa. Therefore, some aspects of the DAT development are discussed in parallel with the modularization process in order to not overlook these interrelations.

Figure 1. A requirements management perspective on the case company’s modularization process. Step 1: Identifying Requirements As a first step in the process of identifying requirements, the case company focuses on the customer’s goals with the DAT and requirements on the modularized product. Module requirements are sorted into four different groups: functional and performance requirements; legislations, building codes and 6

Paper III

standards; usability and process requirements; and business-related requirements and outcome targets. It is difficult to specify all of these requirements before their impact on the solutions can be visualized, which makes it necessary to keep a continuous dialogue with the customer as the work progresses. Continuous dialogue is symbolized with circular arrows in Figure 1. This means that requirements are not static; they evolve as the project progresses. Customer requirements are combined with design requirements found when analyzing the customer’s project portfolio. This analysis aims at understanding the scope of products and finding requirement commonalities on a modular or componential level. Commonalities are found by iterating between viewing technical solutions and deriving their original design requirements. In edge beams, as an example, there are numerous ways of placing and bending reinforcing bars. But if looking from a general perspective, longitudinal bars are meant to deal with transverse loads and brackets are meant to deal with torque forces. So even if all reference projects have different detailed solutions of how to place longitudinal bars and brackets, their purpose are still the same. As suggested by Elgh (2008), checklists are used during this process step. It has in some cases been difficult to ensure that all requirements have been found, despite the use of structured checklists as those suggested in the theoretical framework. This is due to the product variety in the case company’s DAT development projects and the complexity and evolving character of requirements. This is dealt with through continuous customer dialogue, enabling customer feedback on the defined requirements. Requirement identification may be the first step of the process, but the process is far from linear. In the following steps, frequent iterations are made to update the requirement list, based on knowledge developed during the process. Step 2: Formulating Requirements Identified requirements are managed systematically using corporate template files. Intelligent IT-tools enabling requirement traceability, as suggested by Ozkaya and Akin (2007), are at this stage not implemented. Misunderstandings of how requirements should be interpreted sometimes occur, mainly when the customer’s representatives and the case company’s engineers have different knowledge background. These misunderstandings are also sometimes due to unclear interrelations between requirements. In order to decrease the risk of misunderstandings the case company uses a similar approach to QFD (Erixon 1998), though specifying requirement interrelations in list-form instead of matrix-form. These lists do not only serve to specify requirement interrelations, they are also used to prioritize conflicting requirements. The 7


extent of requirements has not yet called for the use of affinity diagrams or classification diagrams as suggested by Erixon (1998) and Elgh (2008). Continuous customer dialogue is, in addition to the structured requirement lists, used to minimize the risk of misunderstandings while formulating requirements. The identification and the formulation of requirements from the customer’s portfolio are based on the engineers’ knowledge and on the communication with the representatives of the customer organization. This is experienced as a complex undertaking which requires much iteration as knowledge of the product evolves. Flexibility in the approach is therefore incorporated by the case company. Step 3: Allocating Requirements to Technical Solutions When allocating requirements to technical solutions, there is a constant iteration between analyzing the functions of solutions specified in the customer’s portfolio and identifying and formulating requirements. The process starts at a general level and gradually moves into more detail. An edge beam itself can be traced back to the function of distributing loads into the ground. On a more detailed level, insulation solutions can be traced back to the function of thermally insulating the foundation. The variety of technical solutions used in the customer’s portfolio to meet these requirements is explored at this process step. Lists and matrices are used to find relations between requirements and technical solutions, both on a modular level and related to module components. Different technical solutions are compared against each other in order to decide which ones should be used in the modules. Though not using DSM or FMT per se, as suggested by Erixon (1998), a similar approach is adopted. In combination with lists from the previous process steps, knowledge is documented to ensure a systematic approach when modularizing the products. This enables traceability of knowledge, as expressed important by Elgh (2008) in Step 2. Focus is on design rationale, i.e. the why and the how (Stokes 2001), which is important for subsequent programming of the DAT (Elgh 2008). During the identification and formulation of requirements and their allocation to technical solutions, the initial effort is based on the knowledge of the engineers at the case company. Progression is however continuously communicated with customer representatives to minimize the risk of misunderstandings and to enable their input. Currently constructability (cf. DFX in Erixon 1998) is not evaluated extensively since personnel such as site managers or construction workers are not included in the process. Production aspects are however further incorporated in Step 6 based on customer feedback.

8

Paper III

Step 4: Generating Concepts Solutions from the customer’s portfolio, suitable for solving the now welldefined requirements, are at this stage chosen, adapted and combined into modules. Deciding which portfolio solutions to use is mainly performed intuitively. The aim is to take advantage of knowledge already built into previous projects, similar to Knowledge Based Engineering described by Stokes (2001). One of the challenges is to take knowledge that is developed as unique solutions in traditional design development and transform it into parametric technical solutions with a built-in variability. In order to ease the work process Excel spead sheets are used to structure the design problem and find correlations between components/module parts, their variability, and defined requirements. This is related to the use of module drivers by Erixon (1998). AutoCAD is used in parallel to visualize possible solutions and explore the design field. The visualizations help the engineers to understand the scope of their suggested designs, thus making it possible to combine solutions to decrease the amount of module-variants. This is based on the commonalities found when identifying requirements. Although the Module Indication Matrix of Erixon (1998) is not used, the case company has adopted a similar, but not as regulated, matrix approach. Flexibility and structure is balanced to facilitate the iterative and creative process. Step 3 and 4 are to a large extent performed in parallel. The outcomes are different though, having a configuration concept as the outcome of Step 4 while Step 3 provides a functional decomposition and a modularization logic. Frequent iterations are also made to Step 1 and 2. The amount of module variability (Erixon 1998) is defined based on a return of investment strategy of the development process of the DAT. In order to ensure that technical requirements regarding e.g. load capacity or deformations are fulfilled, analysis software is used parallel to the modularization process. Module variance is included in these analyses. These analyses are documented in order to enable knowledge traceability (Elgh 2008). Constraints and variant opportunities of the configuration concept are continuously communicated with the customer. It was noticeable at the case company that the first four process steps are highly interrelated, sometimes even difficult to separate. Iterations are frequent enough to question the sequential representation. The sequential representation is however necessary to describe the gradually evolving definitions and the move from requirements to functions to technical solutions to module concepts. It also represents the evolving knowledge of the product.

9


Step 5: Evaluate Concept The last two process steps are performed mainly as part of the development process of the DAT. As described by Erixon (1998), the perspective here is wide. In this case, even wider than the theoretical framework since it incorporates a development process outside of the scope of modularization. The steps are however relevant for the modularization process, since they generate knowledge that is used to refine the module definitions. Due to the wider perspective of these two process steps, they are visualized with a different color than the first four steps in Figure 1. Module variability, both internally within the modules and externally when combining the defined module variants, is evaluated by testing a preliminary version of the DAT against a variety of previous projects. If all these projects can be solved with the DAT, the variability of the modules is sufficient. If not, design iteration is made to further develop the module definitions or, if there is limited value from including the outliers, a redefinition is made regarding the scope of the modularized product. The DAT is also verified through customer tests, based on a well-structured corporate template file. Feedback from this process is used to improve the module definitions and the DAT itself. Step 6: Improve Modules When the DAT has been verified and launched, continuous improvements are made to ensure an up to date and well-functioning tool. These improvements are based on feedback from use of the DAT, including feedback from production. This feedback is used to improve both the DAT itself and the modular product structure defined within the tool. This last process step focuses more on continuous improvements than on final adjustments, thus providing a real-life perspective on the Design for X described by Erixon (1998).

Conclusions Even though the studied modularization process can be described in a sequential manner, the process steps are highly interrelated with frequent iterations and overlaps. The challenge is to find a balance between structure and flexibility, i.e. defining a process to follow that is flexible enough to cope for different modularization and design automation scenarios. The modularization process is also strongly linked to the development process of the DAT, making it difficult to isolate the modularization process. This is particularly visible in the last two process steps, where concept evaluation and module improvements are made as part of the DAT development.

10

Paper III

Both customer and design requirements in the studied modularization process are constantly evolving. Although the customer has an idea of what they require from both the defined product and the DAT at project initiation, these requirements are seldom static throughout the process. The requirement list is continuously updated based on portfolio analysis and communication of modularization results to the customer. Customer communication is integrated as a natural part of the modularization process. Requirement management tools similar to those presented in the theoretical framework are used by the case company engineers to ensure knowledge capture and a structured approach to requirement management. This approach includes lists, matrices and graphical representations, making it possible for the engineers to decompose technical solutions in the customer’s portfolio to find requirement commonalities. These are balanced with the customer’s stated requirements. Though the case company successfully implements modularization, they aim at improving some of their requirement management methods. A requirement management methodology, allowing a balance between flexibility and structure, will be the subject of further research. As will the methodology of finding and using commonalities in house-building projects for enabling product definitions and design automation in an engineer-to-order context.

Acknowledgments We would like to thank the employees at the case company who provided valuable knowledge and information to this research. We would also like to acknowledge the financial support provided by the research project Attract (Attractive and Sustainable Cities in Cold Climates) which is funded by Vinnova (Sweden’s Innovation Agency), HLRC (Hjalmar Lundbohm Research Center) and the participating companies.

References Blecker, T., Abdelkafi, N., Kreutler, G. & Freidrich, G. (2004). Product Configuration Systems: State of the Art, Conceptualization and Extension, Munich Personal RePEc Archive Elgh, F. (2008). “Supporting management and maintenance of manufacturing knowledge in design automation systems” Advanced Engineering Informatics, 22, 445-456. Erixon, G. (1998). “Modular function deployment: A method for product modularization”. PhD Thesis, Royal Institute of Technology.

11

Modularization based on commonalities in house-building requirements Jensen, P., Lidelöw, H. and Olofsson, T. (2015). “Product Configuration in Construction” International Journal of Mass Customisation, 5(1), 73-92. Jørgensen, K.A. (2001). “Product Configuration – Concepts and Methology” Proceedings of the Fourth SMESEM International Conference Aalborg, Denmark 14-16 May Kiviniemi, A. (2005). “Requirements Management Interface to Building Product Models.” CIFE Technical Report #161, Stanford University Lessing, J., Stehn, L., Ekholm, A. (2015). “Industrialised house-building – development and conceptual orientation of the field.” Construction Innovation, 15(3), 378-399. Lidelöw, H., Stehn, L., Lessing, J. and Engström, D. (2015). Industriellt husbyggande, Lund: Studentlitteratur. Malmgren, L. (2010). “Customization of buildings using configuration systems: A study of conditions and opportunities in the Swedish house manufacturing industry”. Licentiate Thesis, Lund University Ozkaya, I. and Akin, Ö. (2007) “Tool support for computer-aided requirement traceability in architectural design: The case of DesignTrack” Automation in Construction 16, 674-684. Pakkanen, J., Juuti, T., Lehtonen, T. (2016, in press) ”Brownfield Process: A method for modular product family development aiming for product configuration” Design studies. Simpson, T.W., Jiao, J.R., Siddique, Z. and Hölrrä-Otto, K. (Editors), (2014). Advances in Product Family and Product Platform Design: Methods & Applications, New York: Springer Stokes, M. (2001). Managing Engineering Knowledge - MOKA: Methodology for Knowledge Based Engineering, ASME Press. Yin, R.K (2014). Case Study Research: Design and Methods. 5 Ed. London: SAGE Zhang, L.L. (2014). “Product configuration: a review of the state-of-the-art and future research” International Journal of Production Research, 52(21), 6381–6398

12

Paper IV

Paper IV

ARKITEKTERS SYN PÅ KRAVHANTERING VID BYGGNADSUTFORMNING: RESULTATREDOVISNING Författarens manuskript Emma Viklund Publicerad som tekniska rapport vid Luleå tekniska universitet (2017) ISSN 1402-1536

Syfte Denna intervjustudie är utförd som ett delmoment i forskningsprojektet Attract WP2.2 Hållbarhetens hus. Intervjuerna syftade till att ge en inblick i arkitekters dagliga arbete med kravhantering samt tankar kring automation, konfigurering och parametrisering. Syftet har inte varit att ge en kvantitativt verifierbar bild av arbetsprocesserna eller att testa redan formulerade teoretiska modeller. Istället var syftet att ge en djupgående förståelse för hur processerna kan gå till i dagens verkliga situationer. Målet med intervjustudien var att lyfta arkitekters perspektiv på att designa för industriellt byggande.

Metod Intervjuerna utfördes i april 2015 med fem arkitekter på fyra olika kontor. Vid ett av intervjutillfällena deltog två arkitekter samtidigt. Arkitekter med olika erfarenhet av att designa för industriellt byggande valdes ut till studien. Urvalet skedde via forskningsgruppens kontaktnät. Några av arkitekterna har lång erfarenhet av att arbeta med entreprenörer med fördefinierade lösningar, och några av arkitekterna har endast stött på industriellt byggande någon enstaka gång. Tre av de intervjuade arkitekterna arbetar främst i förslagshandlingsskeden, en arbetar både i förslagshandlings- och bygghandlingsskeden (men framför allt i förslagshandlingsskeden), och en arbetar med processförbättringar inom företaget. Variationen i urvalet syftade till att belysa olika perspektiv på arkitekters arbetsprocess. För att ge möjlighet att följa upp arkitekternas egna beskrivningar och på så sätt få en djupgående förståelse för arkitekternas upplevelser användes semi-strukturerade intervjuer. En del frågor och diskussionsområden hade förberetts i förväg för att ge en ram för vilka områden som var intressanta att diskutera. Under intervjuernas gång 1

Arkitekters syn på kravhantering vid byggnadsutformning

ställdes även andra frågor och en del frågor omformulerades. Intervjuerna tilläts alltså ta olika form beroende på vad som kom fram i intervjusvaren. Jag tillhandahöll inte några definitioner på de begrepp som förekom i intervjuerna, till exempel kravhantering, plattformar, konfigurering eller parametrisering, eftersom jag ville komma åt arkitekternas egna associationer av dessa. Däremot styrde jag ibland i efterhand diskussionsområdena för att bredda diskussionerna. Alla intervjuer ljudinspelades och transkriberades sedan ord för ord. Med hjälp av det transkriberade materialet kategoriserades svaren i 16 kategorier: 1. Kravställare i husbyggnadsprojekt. 2. Vilka krav som arkitekterna hanterar. 3. Hur kraven presenteras/dokumenteras. 4. Hur karven hanteras. 5. När kraven hanteras i processen. 6. Prioritering av krav. 7. Kontroll av kravuppfyllnad. 8. Vad händer när kraven förändras? 9. Hur arkitekterna upplever att kravhanteringen fungerar. 10. Är det möjligt att kategorisera kraven? 11. Skissar man för hand eller i dator? 12. Samarbetet mellan beställare, arkitekt och entreprenör. 13. I vilka skeden är arkitekten involverad? 14. Tankar kring parametrisering och konfigurering. 15. Tankar om framtiden. 16. Övrigt. Varje kategori sammanställdes sedan var för sig. Som ett första steg i sammanställningsprocessen samlades citat från de olika intervjuerna under respektive kategori. Varje kategori sammanfattades sedan med mina egna ord i kombination med några av citaten, för att ge en samlad bild av ämnesområdena. För att hålla intervjupersoner och företag anonyma i denna resultatredovisning har namn på företag ändrats i citaten. I en del fall har citaten omformulerats så att innebörden av citatet framgår men att företags eller personers namn undvikits. I denna rapport redovisas resultatet från studien. Rapporten innehåller inte någon teoretisk analys av resultatet. Sådan analys kan komma att presenteras i andra publikationer. Rapporten beskriver den samlade bilden av kravhanteringsprocessen, som de fem intervjuade arkitekterna upplever den.

2

Paper IV

Resultat 1. Kravställare i husbyggnadsprojekt Arkitekterna beskriver en samlad bild av vart krav och begränsningar kommer ifrån. Framför allt är det beställaren, entreprenören, kommuner och myndigheter som är kravställare. Olika beställare beskrivs ställa olika krav, där en del beställare är väldigt specifika medan andra beställare inte ställer tillräckligt mycket krav. ”Men det kan ju variera väldigt mycket beroende på vad man har för beställare som jag sa, jag menar det ena extrema är när man har en som vet exakt vad de vill ha. Den andra kan ju vara någon som vill ha ett hus, och sen så är det väldigt mycket upp till oss att styra upp hela processen.”

Brukarens krav hanteras av beställaren eller via marknadsundersökningar. Begränsningarna från entreprenörernas plattformar eller produktionssystem beskrivs också variera beroende på vem som är entreprenör. När det gäller kommuner och myndigheter så nämns framför allt detaljplan, byggregler från BBR och olika typer av miljöcertifieringar (där beställaren ställer krav på certifiering, men certifieringsreglerna ställer krav på byggnaden). I byggprojekt kan det finnas möjlighet att vara med och påverka detaljplaner genom att föra en dialog med kommunen under tiden projektet ritas. En av arkitekterna nämner att det är viktigt att veta markförutsättningarna och vilka krav dessa ställer på konstruktionen. Det nämns också att även arkitekten är en typ av kravställare. De kan ställa krav avseende projektets vision och byggnadsutformningen. 2. Vilka krav som arkitekterna hanterar De krav som ställs på en byggnad beskrivs ha en stor bredd vilket ger en komplex kravbild som kan vara svår att hantera. ”Du kanske tycker att det blir lite luddigt men det är ju, krav är ju liksom allt möjligt, högt som lågt.”

Beställarens krav baseras på vad huset är till för och hur beställaren tänker sig driften. Till exempel kan det vara nyckeltal, antal lägenheter eller lägenhetsstorlekar. ”Men om man tänker bostadshus som jag har ritat, då har det väl oftast varit ganska löst i de här tidiga skedena, de säger ungefär vilken, vilken lägenhetsfördelning de vill ha. Ungefär hur stora de tycker att en två, trea, fyra ska vara. De säger någonting om energikraven och de har förstås kunskaper om vilken typ av parkering de vill ha, om det ska vara garage eller parkering på mark och så.” 3


En av de intervjuade beskriver att beställaren kan sätta sig själv i en fälla genom att ställa för specifika krav, speciellt i tävlingssammanhang. Detta kan leda till att man stirrar sig blind på nyckeltal istället för att se till projektets arkitektoniska kvalitet. ”På Projekt X till exempel, ett projekt jag gjorde, där hade de ju satt krav att man fick ett poäng för varje, man satte en storlek på lägenheterna och sen, för varje kvadratmeter man kunde ta bort i en tvåa då så fick man ett poäng till. Men de förstod ju inte att jag kunde göra tvåor på 31 kvadratmeter för det kan man göra om man tänker till lite. Så att, då hade de satt upp att den som vinner det är den som får mest poäng i slutändan. Istället för att på något sätt tolka in faktorn kvalitet, eller de här lösa värdena som vi arkitekter tävlar ju hela tiden med faktiskt, […].”

Däremot är det inte önskvärt att beställaren ställer för få krav. Genom att begränsa antalet designmöjligheter kan det upplevas lättare att komma fram till en bra slutgiltig lösning. ”[…], jag tycker kanske snarare att många beställare ställer för lite krav.”

En del beställare arbetar med rambeskrivningar med tekniska lösningar som de vill ska användas i varje projekt. En sådan teknisk detalj kan till exempel vara ”att det ska alltid finnas ett takutsprång på minst 300 mm”. Den här typen av krav upplevs inte alltid positiva av arkitekterna, eftersom en sådan teknisk lösning inte nödvändigtvis är den bäst lämpade för alla typer av husdesigner. Andra tekniska krav som kan ställas är till exempel väggarnas U-värden och krav gällande installationer, alltså ”[…] hur huset ska upplevas utifrån komfort och värme, kyla och hela den biten.”

När det gäller miljöcertifieringar beskriver två av de intervjuade att många beställare kräver en specifik certifiering, till exempel miljöbyggnad silver. Men en av arkitekterna menar att det vore bättre att specificera att man vill ha ”någon form utav miljöbyggnad” och att man på så sätt även kan ta hänsyn till vad huset byggs med för material. När det kommer till själva arkitekturen upplever en av de intervjuade att beställaren oftast har förtroende för att arkitekten ska hitta bra lösningar, men det händer även ibland att beställare ställer arkitektoniska krav som till exempel ”anpassad till naturen och, med hög nivå på, eller något sådant där liksom, på arkitekturen […]”.

När brukarvärden hanteras via marknadsundersökningar är det framför allt ”mjuka” värden som hanteras. Viktiga brukarvärden beskrivs vara till exempel: ”Det är ju sådana saker som balkong och parkeringsplatser, läget förstås är alltid nummer ett i alla lägen. Men det handlar om liksom, planlösningar, IT-lösningar, hur mycket kundval, alltså hur mycket man ska kunna påverka i färgval och inredningsval och sådana saker.”

4

Paper IV

Arkitektens roll beskrivs framför allt att vara att värna om projektets vision, vilket har ett ekonomiskt värde för den slutgiltiga produkten. Detta värde kan däremot vara svårt att specificera. Arkitekten har också som roll att värna om den sociala hållbarheten. ”Och samma med sådana här visioner och sådär som kanske byggherrar undrar, men varför gör ni visioner, eller varför har ni sådana saker, varför håller ni på med sådant? Det kostar ju bara pengar, jag vill bara ha nyckeltal, nyckeltal. Jo men det är just för att nå fram till ett resultat och faktiskt också se till den stora affären.”

Begränsningarna som kommer från entreprenörernas plattformar beskrivs bero på vad som är möjligt i deras produktion. Det handlar framförallt om mått, men även en del standardiserade detaljer och tekniska lösningar. Entreprenören är ibland försiktig med att ge för specifika lösningar i sina arkitektmanualer eftersom man vill hålla öppet möjligheten att hitta nya lösningar och fokusera på projektet i stort istället för detaljoptimering i tidiga skeden. ”[…] och jag tror att Entreprenör X har varit lite rädda för att säga att såhär ser våra schaktlösningar ut, punkt. För då har man också förlorat lite att man börjar som arkitekt sitta och rita lösningar där man fokuserar på att hitta schaktoptimeringar. […] alltså det är underordnat på något sätt, för i slutändan så ska de ju sälja en produkt.”

3. Hur kraven presenteras/dokumenteras Kravspecifikationen från beställaren beskrivs sällan vara skriftlig. Oftast diskuterar man vad beställaren vill ha ut från projektet via samtal eller mail. Däremot använder en del beställare referensprojekt eller beskrivningar av färdiga detaljer som de vill ska användas i projekten. Beställarens krav och önskemål samlas i byggnadsprogram och/eller rumsfunktionsprogram. ”Det kan vara att man formulerar skriftligt vad man har för målsättningar, och då kan det ju vara dels sina egna men också att man liksom formulerar kundens krav.” ”Ja, projektets vision tror jag det finns en punkt som i sin tur är ett annat dokument som man ska fylla i tidigt, vilket är just att vi skriver in kundkrav och våra visioner etc. som kan gå parallellt, de behöver ju inte alltid gå, de ska väl inte gå emot varandra för då har man tackat ja till fel jobb så att säga.”

En del entreprenörer använder sig av dokumenterade arkitektmanualer med begränsningar som produktionssystemet innebär för byggnadsdesignen. Detta har blivit vanligare på senare tid. Finns inte arkitektmanualer dokumenterade får arkitekterna själva skaffa sig en övergripande förståelse av produktionssystemets begränsningar.

5


Skissen av byggnaden fungerar, enligt en av arkitekterna, som en sorts samlad bild av alla krav. För att säkerställa att man faktiskt fått med allt i den bilden arbetar man ibland med checklistor. En annan av de intervjuade beskriver att man på dennes kontor har tagit fram ett verktyg för att hantera kraven som ställs i projekten. Detta verktyg är i dokumentform och liknar en checklista, fast med möjlighet att även skriva egna synpunkter. Syftet med verktyget är att säkerställa att alla ser projektet med samma ögon. ”Men det är just det jag har jobbat med senaste tiden väldigt mycket, att ta fram de här verktygen som, som samtidigt blir en sorts checklista, någonting man ser att, man ser samma sak.”

4. Hur kraven hanteras En del önskemål som beställaren har, används mest som riktmärken. Det verkar däremot inte finnas en tydlig formulering från beställarens sida vilka krav som är faktiska skall-krav och vilka som endast är önskemål eller riktmärken. Därför gäller det för arkitekterna att identifiera vilka spelrum de har i projekten, så att man inte lägger tid och energi på ett förslag som man senare måste göra om för att det går utanför ramarna. ”[…] de vet själva att även om de säger att de vill ha 40 % ettor mellan 35 och 40 kvadratmeter så, så funkar det inte riktigt utan det blir någonstans att det är där jag siktar och försöker komma så nära som möjligt.” ”Om man vet, om man har jobbat länge med Beställare X, och vet att man har ett Beställare X-projekt till, då kanske man vill, gillar läget sådär bara, att man inte kan köra precis som man vill. Och så identifierar man vilka spelrum har vi inom det här och så jobbar vi med det, inte med det andra.”

Vilka spelrum man har beror på vilket typ av projekt det handlar om. En del projekt kan fokusera på kostnadseffektivitet medan andra projekt är prestigeprojekt vilket ger större svängrum för att gå utanför ramarna. Prestigeprojekt ger ofta större frihet för arkitekten att komma med nya lösningar, men de innebär också fler diskussioner gällande avsteg från beställarens eventuella kravspecifikation eller rambeskrivning. ”Så att jag tror det är viktigt att arkitekten och de andra konsulterna också har förutsättningarna klart för sig. Att vad är det vi ska bygga här, är det liksom, vilken nivå är det, för det är ju kopplat till vilken intäktsnivå som finns. Självklart. Så att på vissa ställen så kan man göra mer exklusiva saker och på andra ställen kan man inte göra det. Och det är viktigt att man har de kraven klart för sig när man börjar.”

För att säkerställa vad det är som beställaren faktiskt vill ha används skissprocessen. Beställaren får ta del av skisser och idéer genom 6

Paper IV

mellanavstämningar och kan på så sätt kontrollera att projektet rör sig i rätt riktning. ”Men det är så att, då är det nästan så att det är skissprocessen som tar en framåt. Och att det är genom skissprocessen som man tar reda på vad beställaren vill ha.”

Skulle man göra avsteg från beställarens krav kan det i vissa fall innebära långa och krångliga processer för att få avstegen godkända. Arkitekten som citeras nedan ger ett exempel om en beställare som hon/han varit i kontakt med. ”Och sen så får man ju bevaka under hela projektet att på alla sådana ställen där vi har detaljer som har en skärningspunkt med deras rambeskrivningar, att vi kan bevaka dem om vi kan följa dem eller om vi inte kan följa dem och sen så måste man då, om man inte kan följa dem så måste man lämna in en avvikelserapport till dem och få den godkänd av en chef som står högre än deras projektledare. Så det är en ganska lång och jobbig process sådär.”

En del krav som ställs av till exempel kommuner uppfattas inte alltid av arkitekterna som faktiska krav. Det kan bero på att kraven inte är formulerade för att passa alla typer av projekt. En av arkitekterna berättar: ”Sen när det kommer till kommunkrav etc. där är det ju rätt intressant för det är ju aldrig någon byggherre som säger såhär, ja men vi ska följa kommunens nyckeltal när det kommer till p-norm och sådär. Oftast där vill man ju att, vill ju kunden att man faktiskt ifrågasätter det och, för det finns ju väldigt mycket pengar att spara där. […]. Tumma på storleken på ett barnvagnsförråd, utan det är väl såhär en bransch-, alltså gör det bra men sitt inte och rita en massa kvadratmeter för att liksom uppfylla alla dess krav, alltså inom, i en förhöjd nivå, för så blir det ju på något sätt när man läser, man måste ju titta till vilka som bor där, är det bara ettor och det är liksom ett område för ungdomar då är det absurt ifall vi skulle ha liksom plats för 20 barnvagnar, då kanske det räcker med två eller sådär. Det är också rätt outtalat faktiskt.”

Genom att sätta begränsningar på byggnadens arkitektur upplever en del av arkitekterna att det är enklare att komma fram till bra lösningar. Detta är något som entreprenörernas plattformar kan bidra med, om arkitekterna får ta del av dessa i tid. ”[…], om någon säger såhär att kan du berätta en rolig historia för mig, då brukar man liksom såhär, ja vänta nu så ska jag komma på någon. Men säger man kan du berätta en rolig historia för mig som handlar om en norrman, en svensk och en dansk, eller vad man ska säga, då genast så kommer man ju snabbare fram till en rolig historia, även fast man faktiskt stramat in historieberättandet betydligt mindre. Och samma tycker jag det är litegrann med Entreprenör X:s system generellt sätt att man kan nästan snabbare komma fram till bra planlösningar tack vare att det faktiskt finns en begränsning i det.”

7

Arkitekters syn på kravhantering vid byggnadsutformning ”Sen, jag tror inte att det är bra när man mot arkitekten bara, utan att ha några begränsningar, att rita. För då hamnar man ofta i det här att det blir extra allt och så går det inte ihop och så måste man börja skära. Och det man kan skära på när man kommer en bit fram det är liksom, det är kundvärden, och så kanske man skulle ha tänkt helt annorlunda.”

Däremot används entreprenörernas arkitektmanualer mest som ett stöddokument som endast öppnas ibland, enligt två av de intervjuade, i alla fall manualerna från de entreprenörer som arkitekterna är vana att arbeta tillsammans med: ”[…] och jag gör det kanske någon gång i månaden, men det är inte egentligen för att, det är bara för att se proffsig ut mot Entreprenören.”

Ett problem som finns i den komplexa kravbilden är, som tidigare nämnts, att det inte alltid är klart vad som är skall-krav och vad som är bör-krav. En del krav kan också vara svåra att tolka eller ge en viss tolkningsmån. Ett exempel är BBR-kraven där Svensk Standard ger förslag på lösningar som uppfyller kraven men att dessa inte behöver användas utan det är möjligt att hitta andra lösningar. Då måste man däremot kunna bevisa att ens egna lösningar uppfyller kraven, vilket är en ”mycket jobbigare process”. När det gäller begränsningarna från entreprenörernas plattformar får arkitekterna oftast ingen förklaring till varför dessa begränsningar finns. ”[…], det är säkert en resursfråga mycket, men att det blir många gånger så här att arkitekten får en binge dokument liksom, utan att få någon hjälp om hur man ska, hur man ska jobba med de här. […]. För det är, och det är liksom inte rätt sätt, det är inte rätt sätt att ge kraven i den formen, tycker jag. Jag tycker man ska sitta mer och gå igenom tillsammans. Att titta, vi gör såhär på grund av det här och att då också arkitekten kan vara med, med sin kompetens och kanske kan förskjuta vissa saker.”

Kravbildens komplexitet gör att kraven ofta hanteras intuitivt av arkitekterna. Processen kan gå till enligt beskrivningen nedan. ”Oftast om jag ska göra ett projekt med Entreprenör X så handlar det ju om att få till en bra idé och sådär. Och då tittar jag ju oftast på tomten och sen ritar jag ungefär den volym, om det finns krav eller önskemål om en viss mängd volym så ser jag till att få med det i skissen och sen formar jag volymen så som funkar bäst på platsen. Och när jag fått till den volyms-, då tittar jag, är det här, går det här att förena med Entreprenör X:s byggsystem? Är det någonting som liksom, är det för brett hus, är det för smalt hus, och så justerar jag de delarna. Sen går jag in på, oftast ligger ju kravet runtomkring lägenhetsfördelning etc., då börjar jag titta, börjar lite övergripande vilken fördelning på ett normalplan finns, kan jag nå hem här? Ungefär. Och den indelningen i sin tur, alltså, på den skalan tittar jag på, är

8

Paper IV

den kompatibel med Entreprenör X:s system. Så att det, ja, nu kan jag ju inte rita för dig men alltså det, det bygger ju på att hela tiden titta i de lagren.”

Många företag har även egenkontroll-listor som fungerar som en typ av checklistor som man kan använda för att kontrollera att man fått med alla krav. Kravbildens komplexitet och omfattning gör dock att dessa listor sällan används. ”Ja alltså det, det som vi har gjort i vårat kvalitetssystem är att vi har en mängd checklistor som man kan arbeta med om man vill ha stöd i kravhanteringen. Och de utgår ju från ja, plan- och bygglagen, alltså BBR, och hur, ja det kan vara om man jobbar med tidiga skeden med detaljplan så finns det okej, har du kollat vilket koordinatsystem som gäller, har du kollat att tillgängligheten till tomten funkar, har du tittat på lutningar, dutt, dutt, dutt. Finns det liksom, det finns den typen av, eller det är egentligen inte så mycket, jag tror det är väldigt få som använder det i den kreativa processen, men det är som en granskningshjälp. Man har liksom granskningsansvar och kontroller, och innan man lämnar in så måste man ju sätta sig med de här listorna. Det är däremot så att i de flesta fall så tror jag att man inte använder dem så mycket på grund av att det är så extremt omfattande att gå igenom alla de här. Utan man har, med en viss erfarenhet, kunskap om vad det är som är det viktiga i den här typen av projekt.”

Själva designprocessen är en kreativ process där det upplevs vara svårt att införa en strukturerad systematik. ”Det finns ju inte en systematik i det man gör utan det är ett pussel. Vissa är bra på att lägga det pusslet och vissa är bättre på att göra snygga fasader. Men, nej men, det är sådär, det är en analytisk förmåga att liksom ha många faktorer tillsammans och samtidigt göra avvägningar av vad som är ett bra flöde. Och, jag känner inte till att det finns någon metod, annat än att liksom försöka lära sig, förstå, verksamheten. Förstå, och inte bara jobba med någon slags koder eller färger eller sådär, utan man förstår vad som händer. Hur rör man sig, vem går, vilka är det som jobbar där och vad, hur använder man rummen. Och först då kan man ju utveckla de här kraven och få det att bli något riktigt bra.”

En arkitekt beskriver att det finns en kulturskillnad i arbetssätt mellan arkitekter och ingenjörer när det gäller att arbeta fram den slutliga kravbilden, där arkitekternas iterativa process inte alltid verkar uppskattas av ingenjörerna. ”Vi kommer också behöva samarbeta alla och lägga ner mycket jobb på ett förslag och sen förkasta det och sen börja om. Men det är inte så, alltså det är inte den metoden som de flesta andra husingenjörer jobbar, utan de är väldigt vana att få en färdig uppgift, det här mattetalet liksom, man har alla parametrar och sen så räknar jag ut det och sen är det färdigt och sen har de gjort sitt jobb. Och det här att behöva göra om det fem gånger för att vi håller på att försöka forma någonting, det är inte helt självklart. Där kan det, där kan jag tro att de kan tycka att det är frustrerande att arbeta tillsammans med arkitekter eller att arbeta i tidiga skeden.

9


Man vill ha mer tydliga krav, mer tydliga parametrar. Man vill ha ramarna innan man kommer in liksom.”

5. När kraven hanteras i processen För att kunna hantera den komplexa kravbilden hanteras olika krav i olika skeden i processen. Från början har man framför allt med sig de krav som påverkar de grova dragen i byggnadens utformning, till exempel krav från detaljplan, beställarens önskemål om antal lägenheter och lägenhetsstorlekar med mera. Om entreprenör är upphandlad tittar man även på vilka begränsningar entreprenörens byggsystem innebär så att det man skissar på är förenligt med entreprenörens max-mått för spännvidder till exempel. Allt eftersom skisserna utvecklas och förfinas tittar man på krav som rör mer detaljerade saker. Man tittar alltså i lager med krav som påverkar det man ritar. ”[…] man måste sortera bland kraven i de olika lagren så att om man säger så att först gör man väl en grovsortering utav de krav som finns i detaljplan etc., kommunkrav. Och sen så kommer man ju in i ett skede, ett skisskede, där börjar man väl titta litegrann på grunderna till Entreprenören, alltså inte mer i detalj, utan mer övergripande, att det är förenat med deras system. Och sen helt enkelt filtrerar man ner det till, sist kommer ju liksom lagkrav om plushöjder vid entré. Det, men alla ligger ju och verkar, dels ligger de aktivt i olika lager men sen ligger de ju också i, hela tiden i bakhuvudet.”

Det viktigt att alla krav är specificerade redan från början så att man kan ha dem i bakhuvudet medan man jobbar. På så sätt kan man undvika att det dyker upp nya krav som inte är förenliga med den skiss man tagit fram. ”Det är ju direkt liksom, alltså man, vet man vilket system man ska bygga med och så tar man ju med sig de parametrarna i, när man börjar skissa, eller liksom drar första strecket. Så att det, all information som man har, jag menar allt som har med BBR och PBL och alla de här sakerna att göra, det är liksom, det är ju alltid med, det måste ju som in i projektet från början. Och detaljplanen är ju alltid också, vad har vi för begränsningar eller förutsättningar. Och entreprenörens och beställarens önskemål och allting, det är ju liksom, det är det som är grunden när vi sätter oss och börjar. Det är ju direkt egentligen. Så att därför så, därför är det ju bra om man får alla input från början.” ”För det är ju liksom, förutsättningen för att kunna göra rätt från början det är att man har med sig det som ska gälla, det ska inte komma peu en peu en massa nya krav, ojdå liksom, måste vi göra det här också. Då faller saker och ting. Utan att jobba metodiskt och ta reda på det man behöver ta reda på så tidigt som möjligt.”

För att kunna få en effektiv system- och detaljprojekteringsprocess är det viktigt att alla krav är utredda och specificerade innan den projekteringen 10

Paper IV

påbörjas och fler människor börjar arbeta med projektet. Att måsta göra förändringar i det skedet kan bli både dyrt och tidskrävande. ”Sen då när man sätter igång i huvudhandling och bygghandlingar då, då är man ju liksom full bemanning och ska kunna fokusera på att projektera det här och göra, ta fram handlingar som man kan bygga på. Och då är det ju inte bra om man har en massa olösta saker och krav som kommer in i det skedet. För det blir ju dyrt att göra om också då, när jättemånga är engagerade och, så att innan man startar huvudhandlingarna så bör ju alla krav vara utredda och omhändertagna.”

Det kan till exempel bli problematiskt om entreprenören upphandlas för sent i processen, eller att det inte framgår i de tidiga skedena att ett förspecificerat system ska användas i produktionen. ”[…] man relaterar ju alltid till en projektprocess eller tidplan där det finns vissa extremt viktiga hållpunkter. Det är till exempel då detaljplan, ta fram en detaljplan, eller om man ska söka bygglov i tid. Och oftast så synkar ju inte den tidplanen med en upphandlingsplan. Det betyder att vi gör en bygglovshandling där vi har bestämt höjder, våningshöjder, utseende, en hel del typdetaljer, innan man har bestämt vilket byggsystem man ska använda, eller ens vem som ska bygga det.” ”Jag tror att för att det ska bli bra så är det väldigt viktigt att man, om man ska jobba med sånt här, med en plattform, för det, där jag tycker att det blir problem det är nästan alltid när man har jobbat fram ett förslag och kanske redan sökt bygglov. Sen får man veta att det ska vara en plattform. Det är det som är det svåra och det är det som vi oftast sitter med. Att man liksom, det kommer in väldigt viktig information för sent. Och då liksom ligger alla schakten åt fel håll och så ligger precis fel modulmått för att det ska passa med det där, eller, vi måste ändra hela hushöjden för att det var för höga konstruktionstjocklekar i volymelementen eller.”

Som nämnts tidigare kan det ibland vara oklart vilka krav som är skall-krav och vilka krav som är önskemål eller bör-krav. Oklarheten kan dels bero på att det inte är tillräckligt tydligt specificerat men det kan också bero på att det är svårt att i förväg förstå hur kraven påverkar slutprodukten. Därför händer det att kraven förändras sent i processen, när bilden av slutprodukten börjat bli tydlig. ”Och i ett inledande skede så uppfattar man det ju som att det är bindande krav och sen så börjar man närma sig en produkt, man jobbar under lång tid och det börjar hitta sin form. Ja då, då kommer ju alla in och får en tydligare bild av vad som håller på att tas fram och då kan man också se liksom, okej, ja men om vi verkligen ska följa det där kravet 100 procent ut, ja då får det de här och de här konsekvenserna. Vill vi det? Ja sen börjar det ofta liksom flyta lite.”

11


6. Prioritering av krav Designprocessen i tidiga skeden är en iterativ skapande process vilket innebär att man testar sig fram tills man hittat en lösning som anses tillräckligt bra. Den komplexa kravbilden innebär att kraven från olika kravställare ibland kan vara motsägelsefulla. Oftast upplever arkitekterna att detta brukar gå att lösa genom att föra en dialog med de berörda parterna. ”Men det är inte det som är det svåra, utan det svåra när det gäller liksom, krav, det är ju när det finns mer bör-krav eller vill-gärna-krav från byggherre, beställare, eller att man själv har en uppfattning om vad den här byggnaden borde klara av att göra med staden, med rummet, med funktioner. Och där det inte är tydligt att vad är ett krav, utan det är mycket liksom, ja man har höga ambitioner och vill få in många saker och ofta är de motstridiga. Och det är det som är den här skissande processen liksom.” ”Och där, där står väl oftast en liten motsättning ibland mellan vision och fabrikskompatibelt. Och det är ju en dialog som man måste ta fall till fall och detalj till detalj hur viktig den är.” ”Ja, bra fråga för att de krav som ställs från byggherrar till exempel runt ikring lägenhetsfördelningar och sånt det är inte så jäkla viktigt för att de är med i dialogen sen eventuellt ifall de procentsiffrorna försvinner medans, de vet beslutsgången till varför någonting sker.”

I slutändan gäller det att kompromissa för att hitta den lösning som är bäst för projektet. ”Alltså rent, rent praktiskt är det ju en mailväxling där man får försöka förklara för varandra fördelar och nackdelar. Det är ju en dialog som jag tror aldrig någon riktigt går vinnare ur, utan det blir någon kompromiss liksom, jag tror att inom projektet, förhoppningsvis, det är då det blir bra.”

Hur själva processen att hitta en kompromiss går till beskrivs av en av arkitekterna nedan. ”Alltså vissa arkitekter tror jag har någon tanke om att finns inte visionen kvar så, alltså jag försöker hålla mig till det här, och det är egentligen att, är det lathet eller snålhet varför man gör någonting, då blir jag irriterad, men är det för att verkl-, för att projektet faktiskt ska kunna bli av då måste ju vi som arkitekter även förstå att vissa delar måste kompromissas med, och man har kanske satt fel vision ifall man inte kan driva igenom den över huvud taget. Så värderingen där det är nog från projekt till projekt. Ibland märker man att det är, Entreprenör X som kanske ser den enkla lösningen, eller någon ekonomisk besparing för de har satt något pris, men är det det som är en jätteliten detalj som faktiskt hela projektets gestaltningsdel bygger på, då är det ju tvärnej liksom. Så att det, ja det där är fingertoppskänsla skulle jag säga.”

12

Paper IV

Vilka kriterier som gäller för att prioritera krav är svårt att definiera. Men enligt en av de intervjuade har man gjort en typ av prioritering när man väljer att hantera en del krav innan andra under själva designprocessen. ”Alltså det är väldigt svårt att vikta saker, för det är det man gör genom att lägga det i olika lager, det är att man vikt-, har redan där viktat vilket som går först i prioritering. Är det ljus eller är det att man får med ett barnvagnsförråd alltså, […].”

En av arkitekterna beskriver en problematik i att de olika kravställarna har olika kunskap, vilket kan leda till att man inte viktar besluten på samma sätt. ”Det som är svårt är ju att man oftast inte vet prislappen på saker och ting. Det här är ett problem som arkitekt. Och att man inte kan, för då kan inte jag hjälpa till att vikta beslutsgången.”

En annan av de intervjuade beskriver att den här problematiken skulle minska om man arbetade mer tillsammans. ”Vi måste liksom väga intäkter mot utgifter och så titta vad är viktigt på den här platsen, vad ska vi göra här. Så jag tror att jobbar man mera tillsammans så tror jag att kravhanteringen blir enklare.”

En av arkitekterna poängterar också att om en konflikt skulle uppstå (vilket arkitekten aldrig varit med om) i hur kraven mellan beställare och entreprenör ska prioriteras så kommer arkitekten prioritera den som betalar dennes lön. ”Så har det aldrig varit, det har aldrig varit en konflikt. Och skulle det bli det så skulle jag gå på den som betalar min hyra. Om det inte är idioti så att säga. För då har ju någon fel.”

7. Kontroll av kravuppfyllnad När det gäller att kontrollera att kraven faktiskt uppfylls sker arbetet dels med hjälp av egenkontroll-listor men också till stor del intuitivt. Man har även avstämningsmöten med kravställarna där alla parter får möjlighet att kommentera om det är något som saknas eller inte uppfylls i designförslaget. ”Ja alltså vi har ju egenkontroll i våra projekt. Vilket gör att när vi, ofta, aldrig när vi levererar en skiss då men om vi till exempel, oftast till bygghandlingar till Entreprenör X så går ju vi igenom, då har ju vi en egenkontroll där vi checkar av alla punkter som finns och ibland står det ju ’inte relevant och sköts av Entreprenör X i projekteringen’ etc. Det är oftast inga papper som Entreprenör X får utav oss men det är någonting som vi gör för att kvalitetssäkra vår produkt som vi ger ut.” ”Alltså det är ju egentligen en kontinuerlig, man har, alltså avstämningar blir ju, alltså om man nu, när man har ett nära samarbete med entreprenören så blir det ju 13


liksom att ritningarna bollas och man får synpunkter om det är saker som, det här kanske vi, det funkar inte så bra för det blir för lång spännvidd där och bla. bla. liksom. Då får man ju en, det är ju som en kontinuerlig dialog hela tiden. Så det är ju, avstämningarna är ju täta, vilket jag tycker att de ska vara också i det tidiga skedet, för det är ju då man drar upp mycket av riktlinjerna.”

Checklistorna som används för egenkontroll används oftast för att göra en sista koll att alla krav uppfylls. Det poängeras också att det är viktigt att granskningen utförs ordentligt för att undvika att fel och konfliker om ansvar uppstår. ”Mm, ja men det är väl den gyllene regeln att alltid granska lite för mycket. Då minskar väl chansen att det blir fel. Jag har varit i projekt också där man märker att när det varit en dålig granskning och det varit dåliga krav om vem som haft ansvar etc. sen så strular det till.”

Även fast en del krav är väl förankrade i program- och projekteringsskedet händer det att de inte upplevs som fasta krav ute på bygget, vilket kan leda till lösningar som uppkommer under själva produktionen och som inte uppfyller kravspecifikationen. ”Ja svårigheten är väl att, att få alla att veta vilka krav som gäller. Och kanske också en kulturproblematik i att man traditionellt sett varit van att göra lite som man vill. Varje projekt är liksom, börjar man från scratch och det är de personer som sitter där som styr vad som händer och inte. Så att det är mycket en kulturfråga, faktiskt. Som gör att ibland uppfattas det inte som krav för att det är krav men ute i projekten, ute på byggena, så ser man det inte riktigt som krav. Det är lite otydligt. Så det är svårt ända ut i produktionen.

En av de intervjuade arkitekterna berättar att man på dennes företag håller på att införa en rutin där någon utanför projektgruppen granskar projekten och där man kontinuerligt arbetar med avstegshantering. I avstegshanteringen ska det dokumenteras dels vilken typ av avsteg som görs, men också orsaken till varför. På så sätt hoppas man kunna förbättra erfarenhetsåterföringen inom företaget. Med hjälp av denna rutin ska man även kunna hitta den ursprungliga kravbilden för att kunna jämföra vad som var tänkt från början och hur resultatet faktiskt blev. ”Ja om man följer processen och använder verktygen som vi har tagit fram på rätt sätt då kan du hitta den ursprungliga kravbilden som ligger vid olika avstämningspunkter och då kan du jämföra hur det ser ut och hur det såg ut tidigare.”

Även fast man teoretiskt kontrollerar att alla krav är uppfyllda är det sällan man utför mätningar för att kontrollera att detta stämmer efter att byggnaden är uppförd. 14

Paper IV

”Alltså om man nu ska se, så är det ju så att, man säger att man ska klara någonting och entreprenören säger att jo men då bygger vi den här väggen, den klarar det här. Men egentligen, gör den det då? Det kan man ju fundera på. Det är ju inte, alltså, det är ju inte alltid man verifierar i efterhand i projekten vad som faktiskt, om det faktiskt har, det skulle ju vara ganska intressant egentligen att kunna liksom gå och göra mätningar i, egentligen för allting som man har trott att man ska klara. Men det görs väl ibland men det är ju inte i alla projekten man gör det heller.”

8. Vad händer när kraven förändras? Det händer ibland att kravbilden förändras under processens gång. Dessa förändringar kan bero på många saker, till exempel att de initiala kraven inte varit tillräckligt specificerade, att förutsättningar förändras, eller helt enkelt att någon deltagare i processen ändrar sig. En av de intervjuade beskrev ett scenario där arkitekten ändrar sig och kommer med förändringar i projektet. ”Nej men om, det är väl oftast det som är problemet, för då kanske jag kommer med någonting som är fördyrande och då måste det skickas som en ÄTA och sen så har kanske kunden precis, liksom, vatten upp till näsan redan som det är och så. Och så tycker jag ändå att det är så pass viktigt för grundidén eller för projektet i stort så att jag står på mig.”

Att beställaren ändrar sina ställda krav upplevs sällan som ett problem eftersom det är beställaren som får stå för kostnaderna för ändringarna. ”Men när kunden kommer och ändrar, då ändras det. Så är det. Det är väl inget, oftast inget problem. Alltså det är inte så att de ändrar, så blir det på nåt, det är systemhandlinsskede, det är ju till för att ändras. Och det beror ju på vilken ändring det är så att säga, jag ser inga som helst problem med, det är klar att om de vill gå upp en viss storlek med en viss lägenhetsmodul etc. som är, försvårar då den här, man ligger på gränsen i vissa modulmått etc., det är klart man får klura på det.” ”Det beror ju på hur vi har avtalat. Men, men om de betalar för det så antar jag att vi gör det.”

En del förändringar i kravbilden beror på svårigheten att undersöka alla förutsättningar innan man börjar med projektet. Ett exempel på detta är markförutsättningarna som inte alltid är möjliga att ta reda på, eftersom man i ett tidigt skede inte alltid har tillgång till marken. ”Det ska ju inte vara att man upptäcker egentligen att oj, nej fan vi klarar inte det. Men så är det ju ofta. Men helst vill man ju att man ska förstå, att man ska ha grävt i förutsättningarna först så att du vet vad som gäller. Men det går ju inte alltid, ja men du har marken i ett tidigt skede så är det kanske en massa oklarheter på markförutsättningarna […].”

15


När BBR förändras påverkar det också projektens kravbild. Eftersom dessa förändringar inte sker över en natt så upplevs de sällan som ett problem i projekten. ”Nej men då kan man ju säga såhär att när BBR gör förändringar, så sker ju inte det från en dag till en annan. Utan det är ju ofta en övergångsperiod. Och då måste man ju titta på vilka projekt påverkar det här? Och de som påverkas de måste ju då, om de har kommit, det beror ju på hur långt de har kommit men oftast så, så har man kanske inte kommit så långt i ett projekt som, det är ju ofta något år framåt eller så innan nya BBR-kraven, då måste, det måste man ju ta om hand i projektet, de projekt som påverkas så att säga.”

Däremot kan det ibland vara problematiskt att lyckas få in ändrade krav i sena skeden i skissprocessen. ”Det känns som att just de här tillgänglighetskraven de pressas ibland in. De liksom vill man uppnå fast med minsta möjliga marginal. Och när någonting, när något krav flyttas, då är det, då kan det vara svårt att hämta hem de kraven. Storlekar på rum och mått och bredd och längder och vilplan och lutningar och, som är som precis precis anpassade och sen så ska det, och sen ändras det.”

När krav kommer in för sent i processen kan det leda till att man behöver göra stora omtag i projekten för att kunna möta de ändrade förutsättningarna. En av de intervjuade beskriver att detta upplevs extra frustrerande när förändringen inte beror på att förutsättningarna faktiskt har ändrats, utan istället på att information som hade kunnat vara tillgänglig tidigt i processen kommer in till projekten för sent. ”[…] jag kommer tillbaka till de här radhusen vi gjorde och vilken effekt den här husproducentens krav, helt plötsliga krav fick. Dels så skulle de här moduldimensionerna vara mycket högre, det skulle rampas, och för att det skulle bli då så skulle det rampas mycket längre. Följden blir då att tomt-, den här lilla tomtsnutten framför huset eventuellt inte räcker till. Följden blir också att färdig golvhöjd, färdigt golv, ja, kanten mot den sluttande tomten ändras, och det, för att komma ner till gräsmattan i souterrängläget blir det flera trappsteg, då ändras det. Det blir en otrolig så här, dominoeffekt av ett sånt ändrat krav. Och då är det ju ingen kunskaps-, det är ingen kunskap som ändrats eller så, det är bara det.”

9. Hur arkitekterna upplever att kravhanteringen fungerar Överlag verkar arkitekterna reflektera över en del svårigheter som dagens arbetsprocess innebär, men de verkar också acceptera kravhanteringsprocessen som den ser ut idag. ”Så att, sånt dyker ju upp hela tiden och det ställer ju till det med, hur man vill, eller det är väl bara processens gång som jag ser det.”

16

Paper IV

”Svårt att säga, mer eller mindre, det är klart att det inte går att ha hundra procent koll, det tror jag inte, men jag tror att man kan ha tillräckligt bra koll. Men som sagt oförutsägbara saker, det händer ju saker naturligtvis, och speciellt i en planprocess, där går det ju inte att ha allting med sig, för det kommer in krav sent från kommunen, som man kanske inte hade tänkt. Och sen kan det ju vara saker man har glömt bort själv. Men det ska vi ju, det försöker vi ju bli bättre och bättre på. Att få med oss det egna i alla fall.” ”[…] alltså jag tycker aldrig egentligen att kraven är något större problem, alltså det är inte någonting som vi reflekterar, alltså det är bara, det är ju bara en del av vårt jobb att hantera de här olika kraven.”

Kravhanteringsprocessen sker framför allt intuitivt, speciellt när det kommer till hantering av krav som endast diskuteras muntligt. ”Det kan säkert vara så, det kan ju, jag menar det är mänskligt att det liksom, både att vi kan missa att någon har sagt, om den inte, det är sällan att det faller mellan stolarna om du får en manual liksom, eller om du får det på papper. Men om det är någon som har sagt något i förbifarten så är det ju lätt att det, ja men vi skulle vilja det liksom, och så fjutt såhär och sen så kanske det kommer någon liten anteckning eller, från mötet. Men det kanske, om det inte blir tydligt och liksom, så, på så vis så är det ju, jag tycker beställarna kanske skulle kunna vara bättre på att ha tydliga kravspecar, eller, alltså tydligt vad de, vilka krav de ställer.”

När arkitekterna samarbetat med en entreprenör flera gånger hanteras entreprenörens plattform intuitivt, eftersom arkitekten hunnit lära sig plattformens begränsningar. Ett intuitivt arbetssätt innebär däremot en risk för att man missar saker eller gör egna antaganden som kanske inte stämmer överens med vad som faktiskt är specificerat i plattformen. ”Alltså det, där har det ju slappat till sig med åren och det är väl erfarenhet kanske, att man får ett hjul som snurrar och då så vet de att det finns outtalade krav hela tiden. Det är ju kanske också ett av problemen att man inte riktigt alltid vet att det är ett krav eller om det är så att jag förutsätter att det är ett krav.”

Ett intuitivt arbetssätt innebär också att den mänskliga faktorn har en tendens att påverka projekten. En av arkitekterna beskriver att det är viktigt att ha en tydlig struktur i projekten för att undvika att det blir missar. ”Och jag menar det, det kan ju vara missar, det kan vara kommunikation och det kan vara att det, otydlig struktur i projektet liksom. En projektledare som kanske haft för mycket att göra helt enkelt eller så, ja men det kan ju, det är ju mänskliga faktorer liksom och sen så, den som ska samla in den informationen kanske inte har delat ut den till rätt personer vid rätt tid och sådär. […]. Så att, men det är ju ibland också, det kan ju vara projektspecifika saker och det behöver inte alltid vara, det kanske går att göra en manual på det, men det är ändå väldigt mycket enklare om man skriver en tydlig kravspec från det man får till sig. Det skulle jag säga är väldigt 17


enk-, mycket lättare att hantera. En, också en tydlig struktur i projekten och vem är det som tar besluten och liksom, vem ska ge oss så att säga direktiv om vad vi ska göra, så att det är tydligt.”

Även fast det framgår av intervjuerna att en tydlig struktur i projekten är viktigt så upplevs kravhanteringsprocessen svår att strukturera, bland annat på grund av att alla projekt är olika. ”Ja, nej men man måste ju ha förståelse för att om ett Entreprenör X-projekt ska vara optimalt, och optimalt är ju utifrån massa faktorer, optimalt för den kund som ska ha det, optimalt för Entreprenör X:s produktion och optimalt för den process och sådär, så finns det ju massa olika faktorer. Och de är tyvärr inte, det är väldigt svårt att sätta det på ett papper och speca så, men man måste väl ha en övergripande förståelse för att det skiljer sig från projekt till projekt, […].”

Det kan också finnas en poäng med att inte specificera kraven allt för strikt. Hade man gjort det hade man kanske missat möjligheter att utveckla till exempel entreprenörens plattform. ”Annars tycker jag tvärt om att det finns en poäng också, visst, saker och ting ramlar kanske mellan stolarna i vissa krav, eller liksom, fångas upp sent i projektet och hade de varit med från början hade man kunnat lösa det snabbare. Men samtidigt så finns det en, en risk med, […], att det finns en risk med att vara alldeles för kravställd. För vad det gör är att man säkerställer att produkten går igenom produktionen snabbare genom att ställa upp krav. Och det tror jag att Entreprenören förlorar på i längden. För att just när man hela tiden ligger på gränsen, eller man gör lite sådär, det är ju då som Entreprenören faktiskt tvingas ställa om sin produktion och då är det ju sådär, vad ska styra?”

Ett annat problem som poängteras är att kraven ibland specificeras utan att man som arkitekt får veta orsaken till kravet, så att man istället för att ta kravet som ett faktum kan hitta nya lösningar på själva problemet. ”Jo, men det är väl precis det jag sa i början det där också att, just att de inte specificerar sina krav utan att de är outtalade, det är klart att det är ett problem. Speciellt med liksom en del som man har lärt känna som, när man inser att, ibland, de delar inte så jättegärna med sig alla gånger över varför saker och ting är som det är i vissa krav. […], alltså de släpper inte in en på att förstå grunden till problematiken bakom ett krav. Utan det får man fråga sig till. Och ibland glömmer man ju att fråga, utan man tar det som ett fast definitivt krav.”

Att kraven följs utan att ifrågasättas kan leda till lösningar som är onödiga och ger begränsningar i byggnadernas arkitektur. ”Men det kan ju vara sådana här, en rambeskrivning som, nu kommer jag inte ihåg om det var Byggherre X eller någon annan som sa att det ska alltid finnas ett taksprång på minst 300 mm. Och jag förstår tanken, det är en bra tanke. Vi ska fuktskydda våra hus och ett taksprång fuktskyddar övre delen av fasaden väldigt

18

Paper IV

effektivt. Det gör det. Men om vi har till exempel ett 50 meter högt bostadstorn där vi har då ett platt tak, och så ser vi då exempel på att vissa följer den där rambeskrivningen och gör en 30, 300 millimeters takutsprång på den där, som kanske kommer skydda de översta 50 centimetrarna av fasaden men det påverkar hela byggnadens intryck, från att vara en skarpt geometriskt klar geometri, snygg form, till att bli någonting helt annat. […] Visst man ska försöka väga samman alla krav, man ska försöka väga samman alla olika önskemål och liksom så. Men man ska också alltid ifrågasätta alla krav. Det är liksom, vi måste förstå varför, vi ska inte bara göra, följa en dum lista utan att förstå varför, och vad är fördelen med att följa det här kravet, kan vi uppfylla det på ett annat sätt som är bättre?”

Därför upplever arkitekterna att det är viktigt att ifrågasätta och förhålla sig fritt till kraven, för att i slutändan kunna utveckla byggnaden som produkt och på så sätt också utveckla beställarens rambeskrivning och entreprenörens plattform. ”Men vi, vi vet ju nu sen många år att det finns ju en önskan från vissa byggherrar att få allting stöpt i samma form. Men vi vet ju också att det här får ju väldigt stora konsekvenser för den byggda miljön. Det blir ju väldigt likartat. Det, och det kan ibland, om man ska följa deras lösningar så kan det begränsa väldigt mycket i vad man kan göra för utformning, för gestaltning. Och därför så är ju vi, har ju vi ofta, det är inte uttalat, men ofta har vi den där inställningen att vi vet att det där är en lösning som fungerar, om de säger att den fungerar därför att, det blir enklare för dem, då behöver de inte utreda någonting. Men det finns också hundra andra lösningar som fungerar precis lika bra. Och även där så måste ju vi fatta liksom, göra någon slags avvägning, tycker vi att det är värt den här striden att försöka få igenom en avvikande lösning eller inte. Och ganska ofta så försöker vi ju, inte uttallat, men man försöker ändå förhålla sig någorlunda fritt till det där. För att få en produkt som man kan stå för, som man tycker är bra.”

Det är lätt hänt att arkitektoniska värden glöms bort när man jobbar mot effektivisering och processförbättring. Därför poängterar en av arkitekterna att det är viktigt att man inte stirrar sig blind på nyckeltal utan att man även ser det ekonomiska värdet i att bygga god arkitektur. ”Nej men det är väl att de inte riktigt har förstått sig på att titta på ett hus totala kostnad, utan man ser det oftast till en faktisk byggkostnad, faktiskt nyckeltal, BOA, BTA, men de ser inte de ekonomiska vinster kanske som god arkitektur gör genom till exempel en enklare drift, bättre boendemiljö som har ett värde kanske i hyresrättssammanhang. Alltså att de ibland glömmer bort, ja men arkitektoniska värden och inte specificerar dem, specificerar en vision eller så. För det får man inte glömma bort att många utav de här byggherrarna de sitter ju faktiskt på en skyldighet, att deras uppdrag är inte att bygga volym, utan deras uppgift är att bygga kvalitet. Och när det kommer till deras beställning då är den väldigt fokuserad på volym. Det är något lite snett i det.”

19


Beställaren borde därför ställa högre krav på just byggnadens arkitektur, vilket skulle kunna leda till både bättre produkter i slutändan, men också möjligheter att förbättra processer. Varför beställaren i detta fall är viktig är för att beställaren är den som betalar projektet och därför har störst möjlighet att påverka projekt-organisationen, processerna och slutresultatet. ”[…], men jag upplever ändå att det finns, om man säger rent generellt i samhället så är det en, så kanske man skulle våga ställa lite högre krav på byggandet för jag tror ju också att det skulle utveckla processerna och utveckla produkterna om beställarna... För det är ju ändå de som, alltså vi arkitekter kan ju tycka att jo men, det vore ju bra om man kunde göra si eller så, vi brukar ju framföra synpunkter på systemen och sådär, men kraven måste ändå komma från, för att det ska vara intressant att göra förändringar så måste kraven komma från beställarna, om man ska, för att det är ju där pengarna finns. Så att, om det, om man skulle säga ett problem så tycker jag väl i så fall att det är det att man kanske ibland ställer för lite krav som beställare, och så kanske man ser lite för, då blir det som att ekonomin blir, man blir väldigt fokuserad på ekonomin och det är ju bra, det ska man ju vara. Men man måste ju också se liksom det långsiktiga, och, ja. Och då kan det ju vara, den långsiktiga ekonomin och den kortsiktiga liksom, att man väger in mer kanske det långsiktiga också av byggnaderna, och att de ska hålla länge och så, när man väljer material och... Och det är ju ändå beställaren som måste, det är väl de som ska förvalta byggnaden som måste ställa de kraven.”

Orsaken till varför det ofta inte ställs så högra krav på byggnadens arkitektur kan bero på att det är svårt att sätta en faktisk prislapp på arkitektoniska värden. ”[…], det är väldigt svårt att sätta det i ett Excel-ark, och det är väldigt svårt att hävda att, att den här prislappen du har här, har du igen i att du har ett högre söktryck på din bostad eller etc.”

Kravhanteringen kan också bli problematisk av att projekten är långa rent tidsmässigt (i detta fall menade arkitekten hela processen med projekten, inte bara de tidiga skisskedena): ”Det glider iväg och man tänker inte på, och sen ett annat problem är ju att det är långa projekt. Det är ju väldigt många, det är många år och människor kommer och går. De här som börjar med projektet försvinner iväg någonstans och det kommer nya människor under resans gång. Och där tappas ju en massa information. Och därför är det viktigt att man hittar ett enhetligt sätt att dokumentera på där man känner igen sig mellan projekten och man, ja, det ska liksom inte sitta en massa kunskaper i huvudet på någon som sen försvinner.”

Att människor kommer och går kan innebära en del problem också när det gäller hantering av krav som kommer från kommuner. ”Ja det skulle ju vara bra om kommunerna kunde vara lite mer stringenta i sina krav. Det är ofta väldigt personberoende också. Så kommer det en ny handläggare

20

Paper IV

till kommunen som, tycker den någonting annat, för mycket krav är tyckande-saker. Och tyckandet kan komma in precis när som helst. Och det är ju ett problem, speciellt om man tycker olika. En del liksom, den ena tycker på ett sätt och sen kommer den andra och tycker ett helt annat förslag och det har vi ju varit med om i en del projekt. Då får vi rita om hela projektet för att den nya som kommer tycker inte att det där är bra.”

Människor kommer och går även i beställarorganisationen, vilket innebär att man kan få olika svar från beställaren beroende på vart i processen man är. ”Men det som är den stora bristen är ju ofta att man, man sitter ju med helt andra personer i de tidiga skedena. Det är projektutvecklare hos beställaren som är med i den processen, som man bollar med. Och då är det en helt annan agenda. Och så tar man fram en skiss och säljer in den och alla blir, jättebra, spännande, nu kör vi. Sen så flyttas projektet över till produktionsenheten och de kommer med helt andra liksom, där kommer kraven in, alldeles för sent. För ska man jobba smidigt så ska man ju ha en bra samverkan kring de viktiga grejerna redan från början. I de tidiga skedena. Men där finns ju sällan den budgeten, sällan den tiden. Och inte heller det fokuset. Det är ingen som tycker att det är en bra idé att lägga ner en massa tid på att få rätt liksom, modulsystem när vi egentligen bara behöver ta fram en bild som vi ska sälja in hos stadsbyggnadskontoret för att få tillgång till marken.”

Ett liknande problem som arkitekterna stöter på är att många beslut fattas baserat på personliga åsikter och tyckande. Även fast krav finns specificerade så innebär det alltså inte att alla som hanterar dessa krav uppfattar dem på samma sätt. ”Sen tror jag i och för sig att även, det kan även bli problem med bygglov och sånt för där kommer det också in mycket tyckande. Så att det, och det, det är svårt. Sådant där är svårt. Man kanske ska jobba nä-, ja. Nej men det här, det här oförutsägbara tyckandet är ett problem.”

Det händer att entreprenören inte är upphandlar när arkitekten påbörjar sitt arbete. Detta kan leda till att arkitekten behöver designa en byggnad utan att specificera material, vilket leder till svårigheter i senare skeden när entreprenören är upphandlad och dennes byggsystem ska integreras i designen. ”Men som arkitekt är det ganska svårt att tänka, ja men hur, vad, alltså när man inte vet vad det är för material man jobbar med så är det jättesvårt att veta hur man ska tänka med proportioner och det kan vara, balkongerna och allting påverkas ju, utformningen, av vad det är för, är det en betongplatta eller är det trä liksom, jag har ingen aning. Ja då, då blir det väldigt, det blir väldigt svårt att göra arkitektur också då man inte vet vad man jobbar med för material. Så det tyckte jag, det arbetssättet, och det är ju ganska vanligt så att säga. Men sen med samma beställare nästa projekt då där vi gjorde med betongelement då var, det var en partnering ihop med Entreprenör X som entreprenör och där vi hade egentligen en ganska grov skiss när vi satte oss ner från början. Och sen så liksom, ja men det här vill vi bygga 21


i betongelement säger de då, ja men okej liksom, ja men då anpassar man ju efter det och i, har en dialog hela vägen och då, då får man ett bättre resultat tycker jag.”

En annan av arkitekterna beskriver att verktygsmässigt har de nått långt på företaget, men däremot finns det ännu brister i implementeringen. ”Så att vi har kommit långt i rent, alltså, det verktygsmässiga, men vi har implementeringen inte klar. Alltså implementeringen det är den svåraste biten.”

10. Är det möjligt att kategorisera kraven? När man diskuterar kategorisering av krav pratar en av de intervjuade arkitekterna om att kraven hamnar i olika lager som hanteras i olika skeden i processen. Utöver detta ansåg arkitekten att kravbilden är så pass komplex att det skulle vara svårt att göra en kategorisering. På frågan om spontana tankar kring kategorier man skulle kunna dela in kraven i svarade arkitekten: ”Eh, nej det har jag inte. För att det är för många kategorier att ly-, och har en så pass stor komplexitet, plus att jag tror att projekt skiljer sig väldigt mycket åt. Och då kan det vara så, alltså det där handlar väl också om att från projekt till projekt ha en form utav variation i det, där är det ju som sagt vilket utrymme olika projekt har. Hur många krav-kategorier som man ska ta fram. Så är det. Så att nej.”

Två av de andra arkitekterna beskriver en liknande typ av kategorisering i olika nivåer. Den ena beskriver kategorisering efter detaljeringsnivå medan den andra nämner om kraven är skall- eller bör-krav. ”[…], det blir alltid avvägningar, det beror ju på, alltså krav kan ju finnas på många olika nivåer. Det finns ju bindande krav som vi inte kan frångå. Och de, det är ju, på så sätt så blir de ju självklara, att vi måste hantera dem, det kan vara tillgänglighetskrav eller brandkrav eller sådana här saker som man liksom, man måste se till att det går att lösa.”

Arkitekten fortsätter: ”Men alltså att över huvud taget bunta ihop de här olika inputen som formar våra projekt till ett begrepp kravhantering det blir lite förvirrande för oss för det är så olika typer av input. Och kanske är det att vi aldrig har väckt frågan, vi har aldrig tänkt på det på det sättet. Jag tror inte man diskuterar det så i våran bransch över huvud taget. Men kanske skulle det gå att jobba liksom systematiskt med vissa typer av krav, medan vi identifierar vad som är liksom det, det som är, verkligen är krav och vad som är, ja vad ska man kalla det, andra typer av input eller påverkan.”

Diskussionen fortsätter och kommer in på tidsaspekter och detaljeringsgrad som ett kriterium för kategorisering: ”Men det kanske är olika skeden också, för jag menar viss tillgänglighet, det finns några saker såhär som är jätteviktiga i ett tidigt skede. Men alla detaljerna, det kan

22

Paper IV

man ju liksom strunta i så länge man håller på at fortfarande avgöra om det här blir en bra byggnad eller inte, om det blir rätt arkitektur, om den gör rätt med staden, om den liksom har, om det blir rätt mötesplats för människor. Det är på en annan skala liksom, då jobbar vi ju med att forma rummen. Sen så, och då skjuter man ju mycket av det här senare i ett projekteringsskede, att vi vet att vi har såhär många kvadratmeter, självklart går det då få in en handikapptoa. Men vi har inte löst det ännu.”

Två av de intervjuade arkitekterna poängterar att de har svårt att se hur en och samma kategorisering skulle kunna tillämpas på flera projekt, eftersom projekten ofta är väldigt olika varandra. ”Och lite, jag vet inte, det kanske är jag som är motståndare till sådana här pärmar men, papper och checklistor, men lite så har man ju en känsla av att, att göra liksom det här stora förarbetet att försöka tänka ut hela processen i förväg, det är en jättebra övning, en jättebra skola kanske för att lära sig mycket av de här kraven. Men att sen kunna tillämpa just den här metoden man har tänkt ut på ett färdigt projekt, det kommer nästan aldrig funka, för projekten har så olika karaktär. Tror jag.

11. Skissar man för hand eller i dator I de tidiga designskedena använder olika arkitekter olika verktyg för att skissa. En del skissar till största del för hand medan andra använder CADverktyg i stor utsträckning. Utskrivet CAD-material kan även användas som underlag för handskisser. Vilken metod man använder beror också på vad det är man skissar. Skissar man i stora drag upplever en del att det är enklare att skissa för hand medan om man skissar förslag på planlösningar så kan det vara enklare att göra detta i dator. ”Det är i datorn, bara i datorn. Nästan bara, nej, ja med dat-, alltså skissar, alltså, det blir så diffust vad en skiss är för om jag gör en skiss i lägenhetsplaner etc. då gör jag ju det i dator. Men sen finns, finns ju en viss mån av liksom pappersritningar också men de, det är aldrig, jag ritar ju aldrig väggar, väggtjocklekar i, den typen av mer exakta skisser.”

För att kunna presentera ett projekt på ett bra sätt kan det ritas upp som en 3D-modell som man då kan göra renderingar och presentationsbilder ifrån. ”Nej men sen så, oftast så använder vi ju liksom att man, även under skisskedet verkligen bygger upp en komplett 3D-modell. Det är en skiss, vi vet alla att om den accepteras av beställaren så kommer vi i stort sett behöva göra om allt, men vi jobbar ändå på det sättet, att vi bygger upp en komplett modell av huset med fönster och väggar och en hel del detaljering så att vi ska kunna göra 3D-renderingar, fina presentationsbilder för att kunna sälja in förslaget. Då måste man jobba ganska långt i BIM redan i skisskedet.” 23


12. Samarbetet mellan beställare, arkitekt och entreprenör Arkitekterna är slående överens om att det är önskvärt med ett nära samarbete mellan arkitekt och entreprenör. En av arkitekterna berättar att om man skulle ha ett nära samarbete så skulle det vara enklare att hitta lösningar som balanserar arkitektoniska värden med effektiv produktion. ”Sen ritar man igenom någon form av skissförslag. Sen så granskar Entreprenören istället de krav som de ställer på byggnaden. Just för att kringgå helt att jag grottar in mig i det där utan att de då själva granskar handlingarna väldigt noga utifrån dessa krav och sen så skickar de tillbaka istället, när man misslyckas med ett krav, eller vad man ska säga. För på så sätt så kommer de själva att få lägga en vikt i, ja men det där kan vi nog fixa eller etc. Så att den kravspecen ligger hos dem och om de inte får till det så skickar de tillbaka den så får man rätta till det. Kanske. Jag har inte hunnit tänka tanken klart direkt.”

Samma arkitekt fortsätter med att beskriva att en workshop-miljö skulle kunna vara positiv när det gäller de stora dragen i projektet, men att man där inte bör fokusera på alla krav, eftersom kravbilden då blir för stor: ”Framförallt när det kommer till, vad ska Entreprenören leverera och vilken byggherre, vad har de för krav etc. och då att jobba i en workshop för att få fram visionsidén och sådär, Entreprenören kan komma in, jo men vi kan göra såhär för att stärka det, så att det finns, den workshoppen, men det är viktigt tror jag att skilja den typen av krav från BBR-krav, eller, alltså att, för då blir det för mycket och då blir det för torrt.”

En annan arkitekt berättar att ett närmare samarbete skulle kunna minska klyftan mellan arkitektur och produktion: ”Men jag tycker generellt att det, det, det är egentligen så kanske vårat yrke befinner sig lite för långt från bygget i vissa sammanhang. Jag menar om vi visste, hade, hade bra information om hur man vill bygga det här från början. Hade en förståelse för hur man, ja, använde det här systemet så skulle vi kunna jobba mer strömlinjeformat liksom med processen, och det skulle kunna vara bra i många fall. Jag vet det är många sådana här exempel om man tittar på förebilder som man har så har då under många år varit mycket fokus på Schweiz och Österrike, och deras träbyggnads, hur de har utvecklat deras, träbyggnadstraditionen. Och den har ju kommit jättelångt mot ett moderna formspråk, mot bra energieffektiva lösningar och väldigt såhär, högt ansedd tradition som används där nu. Och det bygger ju till jättestor del på att arkitekterna där har fördjupat sig i hur husen ska byggas. Hur, de har ofta liksom lokala, industrialiserat byggande men ganska lokalt där de liksom är med i fabriken och tar fram där, och hjälper till att utforma både husen efter processens förutsättningar men också hjälper till att förbättra processen för att få en bättre produkt.”

Arkitekten fortsätter: 24

Paper IV

”De jobbar väldigt medvetet mot liksom modularisering, mot att få en slimmad process och att, förbättring som är lika med en kostnadsbesparing och en effektivisering. För att vi ofta producerar någon slags ritning och sen så vet vi inte ens om vi ska få vara med i skede två. Eller någon annan tar över.”

En tredje arkitekt beskriver att ett samarbete även skulle kunna ge en positiv effekt för entreprenörens plattform. ”[…] och det tror jag alla gör också att man jobbar med att utveckla sin produkt och det tror jag är väldigt viktigt gentemot, ja men liksom att kunna ta upp hur stora hål man liksom kan ta upp i en vägg och hur, för att det fortfarande ska bli ekonomiskt och, och de, den typen av, där tror jag det är viktigt att, att de, att alltså entreprenörerna inte blir för stela utan att man har liksom en, att man är öppen för, för, att samarbeta och utveckla sin produkt ihop med både då beställare och arkitekter och andra i branschen liksom att man, att man liksom inte fastnar och stagnerar i en produkt som man tycker liksom att det här är bra, utan man måste hela tiden vara med för att hålla konkurrensen.”

Även att det är viktigt att ha ett samarbete tidigt i projekten. ”[…] där är ju liksom alla är ju viktiga i de tidiga skedena, både, att det är ju viktigt att ha en beställare som är engagerad och insatt, och en, i ett partneringprojekt då till exempel där man har en entreprenör med och arkitekt då ibland är det då andra, i tidiga skeden kanske det mest är entreprenören, arkitekten och beställaren som är liksom, eh, och, där, där de har ju liksom sina krav och önskemål som ska liksom, och då, så där blir det ju, det samarbetet i tidiga skedet är ju väldigt viktigt, det är ju där man grundlägger alla kvalitéer, men det är ju oftast ganska strukturerat att man, i, har man ju någon projektledare som styr upp det där oftast.”

En annan av arkitekterna beskriver att man inom dennes kontor försöker få till arbetsgrupper där alla kompetenser är inkluderade vilket man tror kommer ha positiva effekter för arbetet. ”Ja men vi tror att det kommer att förbättra sammanhållningen och förståelsen mellan de olika kompetenserna så att det når längre så att säga. Och så just det här att man sitter tillsammans. […], det underlättar ju väldigt det här att jobba i team, än när folk kommer från olika platser och, och håller på med en massa olika saker. Det, det krävs ju på något sätt att man, man kan jobba fokuserat med ett projekt.”

För att få med sig miljöcertifieringskraven så använder man på ett av kontoret experter som är med i projekten och övervakar att kraven uppfylls. ”Men vi har en process för hur vi jobbar med miljöbyggnad där vi väldigt tidigt startar med en workshop och vi har […] experter som, på detta som är med och. Så att det går att göra rätt och få med sig det. […]. Man kan väl säga att den processen och, att deras deltagande kan bli bättre. Och det är det vi jobbar med nu för att få ihop team bättre som följer projekten eftersom vi ser att ibland liksom släpps lite för löst och sen så plötsligt så går det inte.” 25


Det är också viktigt att de som arbetar med utformningen har kunskap om hur olika utformningar påverkar till exempel miljöbyggnadskraven. Förståelsen för hur olika förändringar påverkar uppfyllnaden av den totala kravbilden skulle kunna öka genom att man har ett nära samarbete mellan olika kompetenser. ”Jo men det är att de, de roller eller kompetenser som krävs är med från början och hela tiden är uppdaterade och känner till tidplaner och känner till när saker och ting förändras. Vi kan då knuffa in, alltså, man kanske gör ibland flera, utan att man förstår vad det får för konsekvenser. Och sen krävs det ju, det är en utbildningsfråga också att liksom, arkitekterna vet vad som förväntas av dem när det gäller miljöbyggnad. Hur ska man, hur ska man, att hålla kontroll på att när man förändrar någon utformning på huset så påverkar det faktiskt vissa saker, men det, och en del är jätteduktiga på det och andra är, har inte hållit på så mycket och är kanske inte det.”

Ett problem som kan uppstå i en traditionell samarbetsform är att det finns ett glapp mellan organisationerna, alltså att man inte har riktigt samma målbild eller att man gör förändringar som inte är förankrade hos dem som i slutändan behöver ta ansvar för dem. ”Men oftast så är det ju så att vi ritar och våran beställare är den som frontar mot, vad heter det, när man har de här tekniska mötena på kommunen, när man ska få allting godkänt. Så att det är de som får ta striden, inte vi. Så oftast så är ju inte de så benägna som vi kanske att göra avsteg. Tekniskt samråd heter det.”

Däremot upplever en av arkitekterna att man som arkitekt ofta får ett stort förtroende från beställaren. ”Så att det, det är ju ofta, och sen när det gäller liksom, om vi säger utformning och materialval och liksom arkitekturen, den tycker jag oftast att vi arkitekter, alltså de, de beställare som vi har haft att göra med så har vi ändå, har vi haft ett, fått stort liksom förtroende att utforma byggnaderna då, och i samarbete då när det har varit ihop med entreprenören och sådär, att de har liksom litat på att man gör, gör bra bedömningar där.”

En annan av arkitekterna beskriver att om man har ett långt samarbete med entreprenören så har arkitekten möjlighet att vara med och påverka den tekniska plattformen: ”Och det är liksom, det tror jag, har man ett nära samarbete med, under längre tid så har man ju möjlighet att vara med och påverka utvecklingen också.”

13. I vilka skeden är arkitekten involverad? En av arkitekterna beskriver att man på deras kontor börjar bli mer och mer involverade i projekterings-processen. 26

Paper IV

På ett av de andra kontoren arbetar man mycket med helhetsåtaganden, att företaget är med från första idén till färdigställande. På ett tredje kontor finns det olika arbetsgrupper men gruppen där den intervjuade arkitekten arbetar jobbar framförallt i tidiga skeden. ”Vi sitter ju ofta i tidiga skeden när man har, ja men ganska skissartad arbetsmetod. Man testar olika sätt och liksom, försöker att testa olika koncept, ritar, och sen så kollar man om det funkar med krav, sen förkastar man, och sen så börjar man om. Men väldigt sådär, iterativ process.”

På det fjärde kontoret är arkitekterna oftast med hela vägen, även i projekteringen. De upplever själva att slutresultatet blir bättre om samarbetet fortlöper genom hela processen. ”Ute i, när det gäller i projekteringen och där att man, ja, alltså vi arkitekter vill ju gärna vara med och tycka till ganska långt in i processen och vissa beställare vill ju också att arkitekten ska vara med hela vägen i processen för att få ett bra slutresultat. Och det är inte alla beställare som förstår att det är viktigt […]. […], det kan ju vara småsaker som, frågor som dyker upp och sådär, som gör ganska stor skillnad. Så därför är det ju viktigt att vi är med i den dialogen hela vägen […]. Så att där kan det ju, där kan man ibland uppleva att det finns lite olika bild av de olika aktörernas roll litegrann och kanske att man kan tycka då att ja men, att ibland kan beställaren tycka att ja men nu, nu har ju ni gjort ert och nu får ju entreprenören ta över här. Och den, det är liksom, i den processen så, där brukar ju vi alltid försöka se till att vi är med ändå och det, oftast har det ju funkat väldigt bra att man har känt att man har lyckats klamra sig fast där i projektet och, nej men de gånger det har blivit bäst resultat det är ju när vi har varit med hela vägen. Och då tycker ju oftast entreprenörerna också att det har blivit jättebra. Så att det är ju litegrann i allas intresse egentligen att det ska bli ett bra slutresultat. För det är ju så att det är ju svårt att, vi har ju tankar och idéer om detaljerna som de kanske inte har förstått. Om man inte får möjlighet att jobba igenom det då tappas ju mycket kanske bort på vägen. Men om man har den möjligheten att vara med i processen, om beställaren, och det är ju, där är det ju beställaren som är egentligen viktig och som måste ställa det kravet på att vi vill att arkitekten är med här, för att godkänna detaljerna och sådär.”

14. Tankar kring parametrisering och konfigurering Genomgående i intervjuerna verkar arkitekternas se en del svårigheter med parametrisering och konfigurering. Arkitekternas egna tolkningar av dessa ords betydelse utgjorde utgångspunkten för intervjusvaren. Det verkar finnas ett motstånd bland arkitekterna mot vad man kan kalla ”arkitekturmaskiner” där alla parametrar är förprogrammerade och den perfekta byggnaden uppstår vid ett knapptryck. Man är rädd att viktiga arkitektoniska värden ska glömmas bort i arbetet mot effektivisering. 27

Arkitekters syn på kravhantering vid byggnadsutformning ”Och det är väl problematiken, alltså det optimala, och jag tror många som tittat inom industriellt byggande kanske också kikar på parametric design eller vad man kallar det, alltså där man egentligen skriver upp alla kravspec, och så kan man ju trycka på en knapp och har man då tillräckligt många parametrar så ska det bli den optimala planen. Det funkar ju uppenbart inte särskilt jättebra. Och det har att göra med att det är så pass många parametrar tror jag som, och så stor komplexitet och sen plus att det ligger någon form av känsla också som är svår att ta på och som är väldigt svår att programmera. Jag tror aldrig att en dator kommer slå en människa. Alltså det är lite där, det, jag kan ha fel här, det kanske blir så, men än så länge är det inte så.” ”Nej för att det var så, det är så, jag råkade, det kan man inte bara tycka är omöjligt, utan det är omöjligt att ha en sådan automatisering. Därför att stadshuset är på östra sidan om det här huset och det här huset står i Östersund så har vi något annat där och… Och vi har vädersstreck som skriftar från, och man har olika krav på olika mötesrum, om man vill ha solen, om man ska ha sovrummen mot öster och vardagsrum åt väster och, så att det, och så vill man bo nära där om man har sovrum där fast man vill gå på toaletten där för att ingen ska se att jag går på toaletten. Och jag vill kunna använda den duschen, fast där är gästrummet och då vill jag inte gå igenom. Jag vill gå och lägga mig tidigt fast jag har besök, eller gäster, och då vill jag inte gå där. Alltså det är ju omöjligt att, alltså det, det är… Så är det.” ”Men det görs ju väldigt mycket lägenheter som man bara delar in den yta som finns i en redan förutbestämd huskropp på ett så bra sätt som möjligt. Men det som är vår ambition och det som var bakgrunden till varför man valde det här yrket var ju att man kanske vill ha någon slags idé om vad det är man gör. Det ska vara idéer som styr. Vad är det för typ av bostad? Vad är det för kvalitéer vi vill uppnå med det här? Vad är det man vill testa eller vad är det man vill göra? Och det vet jag inte om, det känns som att den biten, det här personliga, det personliga angreppssättet eller den personliga val-stilen, eller liksom en Person X-lägenhet har ju sitt signum för de är alltid så bra i detaljerna. Det där försvinner ju liksom.”

För att det ska vara värt att jobba mot standardisering, konfigurering eller parametrisering är det viktigt att säkerställa att de avgränsningar man gör faktiskt ger ett mervärde i slutändan. ”Om det ska vara värt att göra ett sådant här system så måste man ju sätta begränsningar då, och de kanske blir, det blir liksom inte, om du ska standardisera så måste det vara till en liksom, det måste vara värt det. Så det inte bara blir ett klumpigare sätt att jobba på. Men att man jobbar med vissa fördefinierade delar som man får med sig informationen i, det tror jag, och det gör vi ju också, mer och mer. Jag har lite svårt att se att man får in alla krav, men det kanske går. Det vore ju bra.”

28

Paper IV

(Arkitekten syftar på att det är svårt att hitta en balans när det gäller tillräcklig grad av standardisering, eftersom man även behöver kunna vara flexibel för att möta marknaden.) När det gäller konfigurering så finns det en del skepsis mot hur man skulle kunna programmera ett konfigureringsverktyg för byggnadens arkitektur utan att förlora projektens vision, alltså känslan i rummen. ”För att relationen mellan saker och ting, vad är då parametern i det? Är det effektivitet, är det lönsamhet i modulindelningen inom den, eller är det lägenhetskänslan i förhållande till hur utsikten är? Då tycker jag nog att den sistnämnda är det som jag försöker styras efter. Och då går det liksom inte riktigt, det är snarare tvärt om att jag skulle nog se att vi tar en farlig väg att gå.” ”Konfigurera rum skulle, på något vis känns det som att det skulle, kanske inte eliminera möjligheten att tänka framåt men det skulle i alla fall kanske stagnera. Alltså, det effektiviserar, kanske att det skulle effektivisera men det skulle vara svårt liksom, vart tog visionen vägen, vart tar framtiden vägen i sådant fall?”

En av arkitekterna nämner däremot att det finns en del gemensamma nämnare som man oftast tänker på när man ritar till exempel en planlösning. (Pratar om en kollega som jobbar på samma kontor) ”Om vi skulle rita ett vardagsrum i ett bostadshus, då vet jag att vi båda två skulle eftersträva att ha ett vardagsrum i två väderstreck. Till exempel. Vi skulle eftersträva att alla rum skulle ha det. Men i hierarkin så skulle kanske ett sovrum hamna ner i hierarkin. Av det kravet. Det skulle vi båda. Och så skulle vi båda eftersträva att ha kanske våt och kök i anslutning till varandra. Och eftersom att vi båda skulle eftersträva det så kanske det skulle kunna gå att konfigurera det också på det sättet.”

Ett intressant exempel togs upp av en av de intervjuade arkitekterna. Det handlar om att automatiskt generera barnvisor, baserat på analyser av tidigare barnvisor. Detta har visat sig möjligt, men frågan som arkitekten ställer sig är, vad är det man egentligen vill uppnå med detta? ”Men jag, det jag tänkte säga alldeles nyss var att jag hörde ett program på radio som handlade om hur de hade analyserat barnvisor med hjälp av kraftfulla datorer, och hittat liksom strukturer hur periodiserar man, hur hanterar man slingor av, och olika melodisnuttar. Vänta vem var det nu, var det Alice Tegnér, som hade skrivit väldigt många barnvisor. Och sen så lär de det här datorverktyget då att göra en egen barnvisa. Och den följde alla de här reglerna. Och det var, liksom, hade analyserat en stor låtskatt, och så följde den alla reglerna. Och det blev förvånansvärt bra. Ingen hade trott att den här datorn skulle kunna göra en glad trallvänlig barnvisa, men den klarade av det. Och då är frågan, vad är det man har uppnått? Man har kommit så långt att det går, det skulle säkert också gå att få ett datorverktyg som gör planlösningar åt oss. Men frågan är vill vi det? Och vad uppnår vi då? Vad är det vi tappar?” 29


Däremot finns det en del uttryckt nyfikenhet hos några av arkitekterna. Även fast de i de flesta fall verkar skeptiska till den innebörd som de relaterar till orden konfigurering och parametrisering så är de ändå öppna för att undersöka möjligheten till att utveckla dagens arbetssätt. ”Fast jag tycker ändå att det skulle vara intressant att se, tänk om det skulle vara så att det blev bättre?”

När man däremot pratar om BIM verkar arkitekterna ha mest positiva associationer. ”Det är ju till exempel att nu när Revit kommer och det, BIM så att säga, där man kan jobba med hus, vad ska man säga familjer etc. Alltså att inte hitta på hjulet allt för mycket, att programmera till exempel kökslängder så att när jag ritar ett kök i en skiss så redan där så är det kompatibelt med Entreprenören, hur de vill ha sina passbitar etc. […]. Nej men enskilda, alltså som faktiskt går att programmera där man faktiskt kan sätta alla parametrar, som ett kök. Där man faktiskt kan det. Och då kan det generera en snabb, Kökstillverkare X:s system, pang. Och sen kommer det en ful jäkla kryddhylla ja men då kan jag göra en handpåläggning på kryddhyllan istället för att liksom rita alla, och få med alla lagkrav etc. Det tror jag är jättebra.” ”Ja men i vissa fall så skulle man ju kunna använda BIM-verktygen till att på ett enklare sätt hjälpa till med vissa saker. Jag menar vi har ju börjat med sådan att vi har, vi inte bara har dörrar utan nu har vi SIS-dörrar som automatiskt när vi sätter in dem så visar de tillgänglighetsrutorna, zonerna, på båda sidor till exempel. Man skulle kunna jobba på liknande sätt med badrumsenheter, att vi inte bara har själva objektet utan att vi också visar vilka rörelsezoner man har till höger och vänster om en WC-stol för att den ska vara tillgänglig. Sådana här små grejer som skulle snabba på det vanliga ritandet, det vanliga skapandet av olika förslag. Man skulle kunna tänka sig att man fick automatisk redovisning av dagsljusfaktorer till exempel när man satte in fönster, direkt i BIM-modellen utan att behöva göra extra dagsljusberäkningar och mycket sånt där skulle säkert gå, det kommer säkert komma. Där vi får svar på frågor som tar en hel del tid nu, direkt. Men jag tror inte att man ska förvänta sig att vi ska få svaren, men man kanske ska få hjälp med att testa våra svar.” ”Sen allting tror jag blir svårt att, det blir ju som väldigt projektspecifika krav ibland som kanske inte går att lägga in i CAD-programmet på något sätt. Det kanske skulle gå om man har någon som liksom såhär, man klickar i att det och det och det ska vi klara, men vissa saker är ju inte sånt som vi specificerar heller. Jag menar det kan ju vara att vi vill ha viss, att det ska vara det här ventilationssystemet, det är ju inte sådant som vi, vi ritar ju inte in det i vårt, ja men det som har att göra med byggnadens utformning och det, liksom rumslighet, solvärmelast och dagsljus och alla de här sakerna som är liksom, som teknik som baseras på hur man utformar

30

Paper IV

husets rum och, ja. Det, det borde ju gå att kunna lägga in mer, i direkt, i tidigt skede så att man får ut de bitarna. För annars kommer ju det oftast in lite senare att man börjar räkna, titta på det, och då behöver man inte ha någon som sitter och räknar på det speciellt utan då får man ut det.

Däremot beskriver en av arkitekterna att det idag saknas konsekvent implementering av de verktyg som redan finns. En annan av arkitekterna beskriver att BIM kan vara bra, men att det kan vara svårt att överföra för definierade hjälpmedel mellan olika projekt. ”Men jag menar, när vi gick in i det här förskoleprojektet då jag var med och ritade ett antal nya förskolor, då hade ju beställaren inledningsvis väldigt höga ambitioner att de skulle BIM:a på riktigt. De ville få ut allting som gick att få på, alla fördelar man kunde få av att man BIM:ade det här rakt igenom. Och någon sak som de var ute efter, som vi aldrig landade med, det var att det skulle liksom, eftersom de visste så väl hur de sköter en förskola för att den ska fungera, och hur badrum ska fungera, eller en avdelning vad det ska finnas där, så var de lite ute efter att man skulle kunna få hjälp av det i projekteringsverktyget. Vi pratade också olika, vi hade någon konsult här som tog fram förslag på hur man skulle kunna skapa ett rumsobjekt. Så fort man lägger in ett rumsobjekt i ett rum, och säger att det här är en WC, då skulle det automatiskt dyka upp antingen en checklista så att man kunde, man placerade ut handdukskroken, WC, dutt, dutt, dutt, och om de var på rätt ställe i förhållande till varandra så hamnar det bockar liksom i där så är det klart. Eller att man skulle få, vad var det de kallade det, en hink med objekt så att säga, så får du säga att det här är en WC. Då automatiskt placeras det en massa objekt i det rummet. Så att man får med allting. Och sen får man bara placera ut dem själv. Eller, nåt sånt där. Sånt där går att göra. Jag har lite svårt att se att det verkligen skulle underlätta arbetet, för det är så sällan de här kraven verkligen överförs ett på ett. Från ett projekt till ett annat.”

15. Tankar om framtiden För att underlätta i kravhanteringsprocessen beskriver en av arkitekterna att en blandning mellan skriftliga krav och muntliga genomgångar är önskvärt. ”Nej men dels skriftligt så man hinner läsa igenom alla krav som de ställer på byggnaden i dess olika, i någon form av indelning. Men sen också en muntlig genomgång där man kan ifrågasätta om det är ett krav eller en önskan. Så, vilket oftast är en dialog, att ta fram, så man kan vikta saker och ting, som ser bra, funkar bra.”

En av de intervjuade tycker att det optimala vore om begränsningarna som kommer från entreprenörens plattform är tydliga men att de inte påverkar detaljerna i så stor utsträckning.

31

Arkitekters syn på kravhantering vid byggnadsutformning ”Men det är just att nu, det önskescenariot är ju liksom att få den här, en kombination av tydlighet och flexibilitet över det. Men tydlighet när det gäller de stora, vad är liksom konstruktiva delarna, vad har vi för spännvidder och vad klarar vi och vad liksom, vad är, vilken typ av väggar, vad är väggtjocklekar, alla de här viktiga, de är som på något vis... Men sen att ja men okej om man vill att fönstret ska liksom, fönsternischen ska se ut si eller så, där, det tycker jag att man, det vill man som arkitekt gärna kunna vara med och påverka och där tycker jag att det är bra om man var, ha en dialog då i de, ha en flexibilitet i... Så styra men inte detaljstyra skulle jag säga.”

När det gäller önskescenariot för framtida kravhanteringsprocesser trycker arkitekterna på vikten av att inte glömma bort arkitektoniska värden. ”Alltså som jag ser mitt framtida scenario så ser jag egentligen Entreprenören som, nu låter det hårt att säga men, som en jättestor utskrift. En skrivare. Där jag egentligen kan, i framtiden trycka på skriv ut och så går det till Entreprenören som skriver ut det, så att jag har väldigt nära mellan mitt ritbord och den faktiska skrivaren som skriver ut hus. Så ser jag scenariot, inte att det blir en bygg-, en fabrik som på något sätt konstruerar planlösningar.” ”Ja, jag tycker ju att det är oftast kanske för mycket, alltså jag skulle önska ofta att beställaren var mer, ställde högre krav på arkitekturen och ville någonting själva med den och inte släppte för mycket, för, för oss är det ju, det är roligare att ha ett samarbete med en beställare som själv är intresserad och är ett bra bollplank och tycker att det blir viktigt att det blir bra i slutändan och. […]. För nu, ibland kan man ju som känna att det blir mer att det blir fokus mot produktionssystemet och det tekniska systemet och att man ibland glömmer bort arkitekturen. Just för att det är väldigt mycket fokus på pengar och, men att arkitektur behöver ju inte, det behöver ju inte bli dyrare bara för att man lägger mycket tankar, att det finns mycket tankar bakom utformningen, det behöver ju inte innebära att det blir dyrare att bygga utan snarare tvärt om då.”

Även att kommunerna borde se till den stora affären istället för att fokusera på kvadratmeterpris. ”Men sen handlar det väl också om att titta på att, sort-, nej men när man tittar specifikt på Stockholm och närområdena här, de som äger marken som kommunerna, som försöker få ut marknadspriser, de har ju också misslyckats fullständigt i att se den stora affären. Alltså de tittar på x antal kvadratmeter mark och så säger de att det här ska vi ha marknadspris för. Vilket gör att Entreprenören då konkurrerar ju samma som oss, alltså istället för att de sätter ut en spec på att så här mycket kostar marken, visa vad ni kan leverera istället. Vi vill ha riktigt bra boende för unga här. Då börjar man ju tävla med andra faktorer. Vilket gör att, det lägger, man lägger ju inte per automatik på det sättet pengarna i byggherrarnas ficka liksom. Utan man sätter det på kvalitet.”

32

Paper IV

En annan av arkitekterna tar upp att det skulle underlätta i arbetet om fler kompetenser var med redan i tidiga skeden. ”Ja det är klart att man helst vill ha hela bilden klar för sig innan man börjar men det, det kommer ju aldrig att vara så, det, alltså, det är jättekomplext att bygga hus, det är inte som att bygga, vad heter det, bygga bilar på fabrik. Det går liksom inte göra riktigt den jämförelsen, husen bygger man fast ute i våran miljö, de ska stå i hundratals år, de påverkas av så mycket runtikring och omgivningen påverkas av det man bygger och det, så att det måste finnas en viss flexibilitet och en viss tolerans mot att saker och ting är olika. Det tror jag är viktigt. Men ja, men jag tror på att man jobbar i team där liksom, det är inte så att en människa har alla kunskaper för att göra bedömningar för helhetsbilden utan man jobbar tillsammans för den bästa kompromissen. För det är ju det som det handlar om, det är att det är en kompromiss. Men att den ska bli så bra som möjligt. Och att man gör det redan från början och inte skjuter från höften och sen blir det en massa bakslag.”

Det skulle också underlätta om man kunde arbeta med projekten mer koncentrerat, enligt en av arkitekterna. ”Så att det är ju väldigt bra om man har kontroll och att om det nu är i detaljplaneskedet att detaljplanen i alla fall är antagen innan man sätter igång den här projekteringen som kan ju bli ganska effektiv. För jag tror också, det kan bli ganska bra om man kan jobba ganska koncentrerat med det. Och inte att det liksom drar ut jättelångt på tiden och att det blir stopp och man gör något annat, så det blir, det är bättre om man kan göra det där koncentrerat.”

En av arkitekterna beskriver vikten av att verkligen förstå plattformen som ska användas, alltså av att ha ett nära samarbete med entreprenören. ”Ja för det första så måste man ju först-, kunna liksom på ett pedagogiskt sätt förstå och lära sig den plattformen som skall användas. Och ett tätt samarbete där man ges möjlighet till att, man kanske inte bara kallar det avsteg men att an-, ja men att, förstår man det systemet så kan man ju använda det på rätt sätt. Och att i sådant fall då skulle användas på ett sätt som, att man gör bara, att man jobbar bara med det istället för att tampas mot det liksom. […]. Också på något vis, identifiera i sådant fall vilka av de kraven som den här plattformen ställer som är verkligen absoluta. Och att i de kraven samarbeta med arkitekten på ett superskarpt sätt, att de mötena eller de, alltså de materialmötena, de volymskarvarna, de måste vara på det här sättet. Och att då till exempel göra dem så välritade som möjligt. Och kanske i de andra, de andra kraven kan man behandla på ett lite, på ett annorlunda sätt. […]. Men de kraven som verkligen är absoluta, att man vet om det. För det kan också vara någon form av resurseffektivitet att man inte behöver göra avsteg från saker som till slut visar sig vara absolut i alla fall.”

En annan arkitekt beskriver något liknande.

33

Arkitekters syn på kravhantering vid byggnadsutformning ”Det är ju liksom, har man rätt förutsättningar från början, vi går in i det här, vi ska jobba enligt det här plattformssystemet, vi får en utbildning om hur det funkar när man ska projektera det på rätt sätt, lägga in systemlinjerna så att det funkar liksom, så är ju det ett jättespännande sätt att jobba. Tycker jag. Det skulle jag gärna vilja göra.”

Ofta påbörjas processen utan att man har någon direkt plan för hur processen ska gå till, berättar en av arkitekterna. Även detta är något som skulle behöva förbättras. ”Det var ju något som vi pratade om på något studiemöte, att illustrera vår designprocess, till exempel. Men det blir på samma sätt som, vi skulle behöva, man kanske skulle behöva en likadan dialog med en kund som har en väldigt stark uppfattning om hur de vill ha det, så att man liksom får, oftast sätter man ju bara igång utan att ha mycket, tänka så mycket på processen i förväg. Det märker man ju först när det kör ihop sig.”

16. Övrigt En av arkitekterna beskriver ett problem som hon/han ser inom byggSverige: ”Men det stora problemet i svenskt byggande och speciellt med bostäder, det är experiment-lusten som saknas. Det är ett regelsystem som inte uppmuntrar till nytänkande och experimentlust. Specifikt när det kommer till boendet som karaktär. Alltså vi ritar ju faktiskt samma jäkla kärnfamilj än idag som vi gjorde på 20-talet, så planlösningar från 20-talet funkar lika bra för våra behov som, eller de kanske de inte gör, de kanske inte funkar lika bra för våra behov, men lägenheterna som produceras ser likadana ut. Och där har man ju satt, och den här efterfrågan som finns, på något sätt också satt något filter där. Det i kombination med den här att pengar styr, fast det blir fel att säga det, för det är klart att pengar styr. Men det är ju svårt att, alltså det där är det väl som arkitekt, att vara duktig på att argumentera för ekonomiska värden. Så det får man ta på sig själv snarare än att ropa att man vill ha en större stol att sitta på i byggbranschen.”

En annan av arkitekterna poängterar att kravhanteringsprocessen förmodligen ser annorlunda ut om man jobbar i projekteringsprocessen istället för i tidiga skeden: ”Nej men det jag skulle vilja lägga till det är att om du pratar med projektörer till exempel, som sitter hela dagarna i en projekteringsprocess, så får du ju helt, kanske helt andra svar. För att deras brottningsmatcher med krav ser helt annorlunda ut än våra.”

34

Paper IV

Diskussion Denna resultatpresentation redovisar fem arkitekters syn på kravhantering i deras dagliga arbete med att utforma olika typer av byggnader. Framför allt har utformning av bostäder diskuterats, men även andra byggnader som till exempel förskolor har nämnts i intervjuerna. Arkitekterna har olika erfarenhetsbakgrunder, några har arbetat länge i branschen och några har endast arbetat i några år som arkitekt. Flera olika entreprenörer och beställare diskuterades i intervjuerna, beroende på arkitekternas erfarenheter. Dessa beställare och entreprenörer använder sig av olika typer av standardiseringar i plattformar. Till exempel har en del beställare inga rambeskrivningar alls medan andra beställare har standardiserade tekniska lösningar som de vill återanvända i projekten. Entreprenörerna som diskuterats har till exempel preferenser på produktionsmetoder, använder sig av förtillverkade element, eller producerar byggnadsvolymer som monteras på byggarbetsplatsen. Denna spännvidd i arkitekternas erfarenheter och i de exempel som arkitekterna tagit upp under intervjuerna gör att en jämförelse mellan de olika intervjuerna är svår. Det har däremot varit möjligt att sammanfatta arkitekternas erfarenheter i de 16 kategorierna som presenteras i denna rapport. Bredden på innehållet i dessa sammanfattningar kan ses som en styrka i studien, då flera relevanta reflektioner från arkitekternas sida har kunnat belysas. Hela intervjuempirin har använts till resultatsammanställningen, däremot redovisas materialet i en kondenserad form för att ge en tydligare överskådlighet. Eftersom inga av begreppen som användes i intervjuerna förklarades i förväg (t.ex. kravhantering, plattformar, parametrisering, konfigurering m.fl.) kan en potentiell felkälla tänkas vara att olika respondenter tolkar dessa begrepp olika. Intervjuernas karaktär, det vill säga att respondenterna tilläts föra långa resonemang och att följdfrågor ställdes för att säkerställa att jag som intervjuare förstått vad respondenterna menat med sina svar, bidrar dock till att minska risken för feltolkningar. För att minska risken för feltolkningar vid redovisningen av resultatet kompletteras mina beskrivningar med citat. Läsaren ges på så sätt möjlighet att bilda sig en egen uppfattning om empirins innebörd. Citaten valdes ut för att spegla de olika synvinklarna på respektive presenterad kategori. Stort tack till alla som bidragit till studien med sina erfarenheter och reflektioner.

35