RICS Research RICS Research RICS Research
nD modelling – a driver or enabler for construction improvement?
RICS Research paper series Volume 5, Number 6 April 2005 Ghassan Aouad, Rachel Cooper, Changfeng Fu, Angela Lee, Amanda Marshall-Ponting, Joseph Tah, Song Wu University of Salford
Dr Angela Lee is a Research Fellow at the University of Salford. She leads and works on several research projects including ‘From 3D to nD Modelling’, ‘nD Game’, ‘PeBBu Thematic Network’ and ‘Revaluing Construction’. Her research interest lies in performance measurement, process management, performancebased building and process and product modelling. She has published extensively in both journal and conference papers in these fields. She completed a BA (Hons) in Architecture from the University of Sheffield and has a PhD from the University of Salford.
The 3D to nD Modelling Project at the University of Salford is led by Professor Ghassan Aouad, who is Head of the only 6* rated UK university School of Construction and Property Management. He is also Director of the EPSRC funded Salford Centre for Research and Innovation (SCRI). The 3D to nD project brings together a focussed research team from the University of Salford, comprising Professor Rachel Cooper, Professor Joseph Tah, Dr Song Wu, Ms Amanda Marshall-Ponting, Dr Angela Lee and Mr Changfeng Fu, who are endeavouring in the project to bring about the widespread adoption and use of nD-enabled construction.
April 2005 Volume 5, Number 6
nD modelling - a driver or enabler for construction improvement?
Angela Lee, Song Wu, Amanda Marshall-Ponting, Ghassan Aouad, Joseph Tah, Rachel Cooper and Changfeng Fu School of Construction and Property Management, University of Salford United Kingdom
© RICS April 2005 ISSN 1464-648X
Aims and scope of the RICS Research Paper Series
The RICS Research Paper Series Editor
ISBN 184 2192 000 The aim of the RICS Research Published by
Paper Series is to provide an outlet for the results of research relevant
RICS
to the surveying profession.
12 Great George Street
Papers range from fundamental
London SW1P 3AD, UK
research work through to
[email protected]
innovative practical applications of new and interesting ideas. Papers
The views expressed by the
combine academic rigour with an
author(s) are not necessarily those
emphasis on the implications in
of the RICS nor any body
practice of the material presented.
conneected with it. Neither the author(s), nor the RICS accept any
For more information on the
liability arising from the use of this
RICS Research Paper Series,
publication.
please contact:
Copies of this report can be made
Stephen Brown
free of charge for teaching and
Head of Research
research purposes, provided that
RICS 12 Great George Street
• •
the permission of the RICS is
London SW1P 3AD
sought in advance
United Kingdom
the copies are not
[email protected]
subsequently resold •
the RICS is acknowledged
Tel: +44 (0)20 7695 1568 Fax: +44 (0)20 7334 3722
Professor Les Ruddock School of Construction and Property Management University of Salford Salford Lancs M7 1NU United Kingdom Tel: +44 (0) 161 295 4208 Fax: +44 (0) 161 295 5011 Email:
[email protected]
Panel of referees Akintole Akintoye Glasgow Caledonian University Scotland Ghassan Aouad University of Salford England David Baldry University of Salford England Malcolm Bell Leeds Metropolitan University England Alistair Blyth University of Westminster England Thomas Bohn University of Leipzig Germany Paul Bowen University of Cape Town South Africa Terry Boyd Queensland University of Technology Australia David Chapman University College London England Charles Cowap Harper Adams University College England Hoffie Cruywagen University of Pretoria South Africa Tim Dixon College of Estate Management England Christopher Fortune Heriot-Watt University Scotland Andy Hamilton University of Salford England Barbara Haward University of Portsmouth England
Michael Hoxley Anglia Polytechnic University England Will Hughes University of Reading England Eddie Hui Hong Kong Polytechnic University Hong Kong Norman Hutchison University of Aberdeen Scotland Ramin Keivani Oxford Brookes University England Richard Laing Robert Gordon’s University Scotland S M Lo City University of Hong Kong Hong Kong David Lowe University of Manchester England William McCluskey University of Ulster Northern Ireland John Mansfield Nottingham Trent University England Jacob Opadeyi The University of the West Indies Trinidad and Tobago Rob Pickard Northumbria University England David Proverbs Wolverhampton University England Rainer Schulz University of Aberdeen Scotland Martin Sexton University of Salford England Low Sui Pheng National University of Singapore Singapore Thomas Uher University of New South Wales Australia Clive Warren University of Queensland Australia Christine Whitehead London School of Economics England Peter Wyatt University of the West of England England
Contents
Introduction
7
From 3D to nD modelling
8
Methodology
16
Workshop findings
17
Conclusion
38
References
41
nD modelling - a driver or enabler for construction improvement? Angela Lee, Song Wu, Amanda Marshall-Ponting, Ghassan Aouad, Joseph Tah, Rachel Cooper and Changfeng Fu University of Salford
Abstract The terms ‘nD modelling’ and ‘nD CAD’ are becoming escalating idioms associated with ICT-based (information communication technologies) building design; a concept which has been heightened by the £0.5 million funded 3D to nD modelling project at the University of Salford. An nD model is an extension of the building information model that incorporates multi-aspects of design information required at each stage of the lifecycle of a building facility. This paper will explore the concept of nD modelling, lay out the nD modelling opportunities and barriers for implementation within a business context, and conclude by examining whether it can be applied industry-wide or remain a PR tool for the big players in the industry. The issues surrounding nD-enabled construction are not just technological, but also and cultural, human and process related. The paper suggests that the nD modelling approach is ideal, but not achievable in the short term, largely due to the industry’s unwillingness to work together in an interdisciplinary way.
Contact Angela Lee School of Construction and Property Management University of Salford Bridgewater Building Meadow Road Salford Lancs M7 1NU United Kimgdom Tel: +44 (0)161 295 5855 Email:
[email protected]
nD modelling - a driver or enabler for construction improvement
Introduction
o
ne of the main challenges facing the construction industry today is how to improve the efficiency and effectiveness of the integrated design and construction process. Moreover, what contribution can the effective use of information technology make to this? A concept that has recently been forefronted as a solution to solve the ills of construction is nD modelling/ nD CAD. An nD model is an extension of the building information model, which incorporates multiaspects of design information required at each stage of the lifecycle of a building facility (Lee et al, 2003). nD is unlike traditional 2D and 3D CAD systems whereby the building design is represented in multiple files made up of lines, arcs and circles, and the building information is contained within several document formats such as spreadsheets and word processing applications. The nD database is constructed with intelligent ‘objects’ which represent building elements like walls, doors and windows. From this central database, different views of the information can be generated automatically; views that correspond to traditional design documents such as plans, sections, elevations, schedules etc. As the documents are derived from the same central database, they are all coordinated and accurate – any design changes made in the central model will be automatically reflected in the resultant drawings, ensuring a complete and consistent set of documentation (Graphisoft, 2003).
the nD modelling opportunities and barriers for implementation, and concludes by examining whether it can be a practical proposition with industry wide applicability or whether it’s merely a buzzword destined to remain a PR tool for the big players in the industry.
One of the main challenges facing the construction industry today is how to improve the efficiency and effectiveness of the integrated design and construction process.
This paper explores the concept of nD modelling, presents the current development of the 3D to nD modelling project, lays out www.rics.org
RICS Research
7
From 3D to nD modelling
2D and 3D modelling/ CAD 2D and 3D modelling/ CAD in the construction industry takes its precedence from the laws governing the positioning and dimensions of a point or object in physics, whereby a three number vector represents a point in space: the x and y axes describing the planar state and the z axis depicting the height (Lee et al, 2003). However, 3D modelling in construction goes beyond the object’s geometric dimensions and replicates visual attributes such as colour and texture. This visualisation is a common attribute of many AEC design packages, such as 3D Studio Max and ArchiCAD, which enable the simulation of reality in all its aspects or allow a rehearsal medium for strategic planning.
4D modelling/ CAD Combining time sequencing and visual environments with the 3D geometric model (x, y, z) is commonly referred to as 4D CAD (Rischmoller et al, 2000). Using 4D modelling/ CAD, the processes of building construction can be demonstrated before any real construction activities occur (Kunz et al, 2002). This will aid the identification of possible mistakes and conflicts at the early stage of a construction project, and enable stakeholders to predict the construction schedule. 4D modelling/ CAD is the subject of many research programmes around the world and commercial packages have started to emerge. In the USA, the Centre of Integrated Facility Engineering (CIFE) at Stanford University has implemented the 8
RICS Research
concept of the 4D model on the Walt Disney Concert Hall project. In the UK, the University of Teesside’s VIRCON project integrates a comprehensive core database designed with Standard Classification Methods (Uniclass) with a CAD package (AutoCAD 2000), a Project Management Package (MS Project) and Graphical User Interfaces as a 4D/ VR model to simulate construction processes of an £8 million, 3 storey development for the University’s Health School (Dawood et al, 2002). Commercial packages are also now available, such as 4D Simulation from VirtualStep, Schedule Simulator from Bentley, and 4D CAD System from JGC Corporation.
nD modelling nD modelling builds upon the concept of 4D modelling by further integrating an nth number of design dimensions into a holistic model, thus enabling users to portray and visually project the building design over its complete lifecycle (Lee et al, 2003). nD modelling is based upon the building information model (BIM), a concept first introduced in the 1970s and the basis of considerable research in construction IT ever since. A BIM is a computer model database of building design information, which may also contain information about the building’s construction, management, operations and maintenance (Graphisoft, 2003). The idea evolved with the introduction of object oriented CAD; the ‘objects’ in these CAD systems (e.g. doors, walls, windows, roofs) can also store non-graphical data about the building in a logical structure. The BIM is a
nD modelling - a driver or enabler for construction improvement?
repository that stores all the data ‘objects’ with each object being described only once. Both graphical and non-graphical documents, such as drawings and specifications, schedules and other data respectively, are included. Changes to each item are made in only one place and so each project participant sees the same information in the repository. By handling project documentation in this way, communication problems that slow down projects and increase costs can be greatly reduced (Cyon Research, 2003). Leading CAD vendors such as AutoDesk, Bentley and Graphisoft have started to heavily promote BIM. However, as these solutions are based on different, noncompatible standards, an open and neutral data format is required to ensure data compatibility across the different applications. Industry Foundation Classes (IFCs), developed by the International Alliance for Interoperability (IAI), can provide such capabilities. IFCs provide a set of rules and protocols that determine how the data representing the building in the model are defined and the agreed specification of classes of components enables the development of a common language for construction (IAI, 2004). IFCbased objects allow project models to be shared whilst allowing each profession to define its own view of the objects contained in that model. This leads to improved efficiency in cost estimating, building services design, construction, and facility management: IFCs enable interoperability between the various AEC/FM software applications allowing software developers to www.rics.org
use IFCs to create applications that use universal objects based on the IFC specification. Currently, most of the major CAD systems provide support for IFC export and this will ensure that the data is consistent and co-ordinated. Furthermore, this shared data can continue to evolve after the design phase and throughout the construction and occupation of the building.
nD modelling tool prototype development The 3D to nD research project at the University of Salford is developing a holistic nD modelling tool using IFCs, to help improve the decision-making process and construction performance by enabling true ‘what-if’ analysis to be performed to demonstrate the real cost in terms of the variables of the design issues (see Figure 1). Therefore, the trade-offs between the parameters can be clearly envisaged, which will help to: • • • • • • • •
Predict and plan the construction process Determine cost options Maximise sustainability Investigate energy requirements Examine people’s accessibility Determine maintenance needs Incorporate crime deterrent features Examine the building’s acoustics
The objective of the nD modelling tool is to enable an informed design decision based on a variety of design perspectives. Traditionally, a whole host of construction specialists are involved in instigating the design of modern
RICS Research
9
buildings. With so much information and from so many experts, it becomes very difficult for the client to visualise the design, any changes applied, and subsequent impacts on the time and cost of the construction project. Changing and adapting the design, planning schedules and cost estimates to aid client decision-making can be laborious, time consuming and costly. Each of the design parameters that the stakeholders seek to consider will have a host of social, economic and legislative constraints that may be in conflict with one another. Furthermore, as each of these factors vary – in the amount and type of impacts they can have – they will have a direct impact on the time and cost of
the construction project. The criteria for successful design therefore will include a measure of the extent to which all these factors can be co-ordinated and mutually satisfied to meet the expectations of all the parties involved. Specialist design criteria input is usually undertaken in a sequential step-by-step fashion whereby the design undergoes a number of changes; after satisfying the legal requirements, it then proceeds to the next consultant who in turn makes a number of design recommendation changes. Design changes are made in isolation from each other in an over-the-wall manner, whereby
Changing and adapting the design, planning schedules and cost estimates to aid client decision-making can be laborious, time consuming and costly.
10
RICS Research
nD modelling - a driver or enabler for construction improvement?
Figure 1: What-if analysis indexes of the 3D to nD modelling project
each discrete change plays little or no regard to the next (see Figure 2). Therefore, it is often difficult to balance the design between aesthetics, ecology and economism – a three dimensional view of design that
acknowledges its social, environmental and economic roles – in order to satisfy the needs of all the stakeholders. These problems are mainly attributed to the vast amount of information and knowledge that is required
Figure 2: Sequential over the ‘brick wall’ approach to design and construction
www.rics.org
RICS Research
11
to bring about good design and construction co-ordination and communication within a traditionally fragmented supply-chain. The complexity of the problem increases with the fact that this information is produced by a number of construction professionals of different backgrounds. Therefore, without effective implementation of IT and processes to control and manage this information, the problem will only intensify as construction projects become more and more complex, and as stakeholders increasingly enquire about the performance of buildings (sustainability, accessibility, acoustic, energy,
maintainability, crime etc). The nD prototype’s key feature is its ability to incorporate numerous design perspectives in one system, and its subsequent capacity to systematically assess and compare the strengths and weaknesses of different design scenarios presented by the nD knowledge base. The system architecture is illustrated in Figure 3. The tool builds on the concept of BIM, and is IFC-based: • nD knowledge base: a platform that provides information analysis services for
Figure 3: System architecture of the nD modelling prototype tool
12
RICS Research
nD modelling - a driver or enabler for construction improvement?
www.rics.org
RICS Research
13
Figure 4: Screenshot of nD prototype tool, displaying AHP technique for determining suitable building material in accordance to lifecycle costing, acoustics and environmental impact factors
Figure 5: Screenshot of nD prototype tool, for determining lifecycle cost options using the MCDA technique
14
RICS Research
nD modelling - a driver or enabler for construction improvement?
the design knowledge related to the various design perspective constraints of the nD modelling (i.e. accessibility requirements, crime deterrent measures, sustainability requirements etc). Information from various design handbooks and guidelines on the legislative specifications of building component will be used together with physical building data from building information model to perform individual analysis • Decision support: multi-criteria decision analysis (MCDA) techniques have been adopted for the combined assessment of qualitative criteria (i.e. criteria from the Building Regulations and British Standard documents that cannot be directly measured against in their present form) and quantitative criteria (e.g. expressed in geometric dimensions, monetary units etc). Analytic Hierarchy Process (AHP) is used to assess both qualitative criteria (i.e. criteria that cannot be directly measured) and quantitative criteria (e.g. expressed in dimensions, monetary units, etc) So far, the nD prototype tool (see Figures 4 and 5) incorporates whole-lifecycle costing (using data generated by Salford’s Life-Cycle costing research project), acoustics (using the Rw weighted sound reduction index) and environmental impact data (using BRE’s ‘Green Guide to Specification’ data).
www.rics.org
RICS Research
15
Methodology
A series of workshops were conducted to gain consensus a the boundaries of the nD modelling concept. The workshops were used to define, develop and validate the proposition of nD modelling project over a period of 18 months and included an academic research team workshop, a national and an international workshop. The first academic workshop set about defining the need and scope of nD modelling using a case study exemplar. To identify the similarities and differences in design requirements between various disciplines, a provisional plan of an office space for 10 researchers was circulated with instructions that improvements and developments should be made to the design based on individuals’ disciplines (client, enduser, accessibility, sustainability, maintainability, maintainability, crime, acoustics and energy). Various designs were suggested which clearly emonstrated that it was extremely difficult to accommodate all stakeholders during the design process. It also drew attention to the problem of decision trade-offs between conflicting design issues. For instance, from a crime prevention perspective, windows should be small, open inwards and positioned near the ceiling to reduce intruder access whereas from an access perspective, windows should be large, glare-free and positioned lower to enable a wide range of users to operate it. The purpose of the national workshop was to gain support for nD modelling and to ensure that these were in line with the demands and requirements of the UK’s wider academic and industrial communities. The attendees 16
RICS Research
included contractors, clients, suppliers and architects, ranging from one person organisations to large multi-national organisations. The spectrum of participants was to reduce any inherent exclusion of industry players in the development of nD, and to gain interest and acceptance of the work. The aims were to be achieved by asking the group to discuss three questions: ‘how may IT support integrated design and construction in the future, in both 5 and 20 years time’; and ‘what are the barriers and opportunities for implementation?’ Finally, the international workshop brought together leading players from around the world, again from both industry and academia. It was envisaged that this event would augment the findings from the previous workshops. The detailed findings of the academic workshops have been reported elsewhere – see Lee et al (2003) for further details and additional references. The workshops of most relevance to the question of the practicality of the nD modelling vision’s propositions are the national and international workshops as these are focussed, not upon the vision of the ‘ideal’ construction industry, but upon the practicality of that vision’s propositions in dealing with the problems and issues that the industry stakeholders actually experience. It is consideration of these contextual aspects – the opportunities, and barriers to implementation – which will determine whether the technology will eventually be taken up.
