Classification Framework for Context Data from Business Processes Michael Möhring1, Rainer Schmidt2, Ralf-Christian Härting1, Florian Bär2, Alfred Zimmermann3 1Aalen University University of Applied Sciences 3Reutlingen University
[email protected] [email protected] [email protected], 2Munich
Abstract. New business concepts such as Enterprise 2.0 foster the use of social software in enterprises. Especially social production significantly increases the amount of data in the context of business processes. Unfortunately, these data are still an unearthed treasure in many enterprises. Due to advances in data processing such as Big Data, the exploitation of context data becomes feasible. To provide a foundation for the methodical exploitation of context data, this paper introduces a classification, based on two classes, intrinsic and extrinsic data.
1
Introduction
Enterprise 2.0 is a business concept used by more and more enterprises and organizations in order to quickly adapt to changing customer requirements and to exploit the innovative potential of customers and suppliers [1]. A core concept of Enterprise 2.0 is the use of social production [2] in addition to classical tayloristic [3] organizational approaches. By using social software employees are empowered to create data in the context of business processes, that means outside of predefined procedures and transactional information systems such as ERP systems [4]. This context data is mostly semi- and unstructured. It is estimated that more than 85% of all relevant business data are unstructured [5]. Unfortunately these context data is a still unearthed treasure in many companies. Although information systems provide very elaborate means for managing structured information such as accounting and transactions, such a broad processing support lacks for context data. The reason is, that, context data are difficult to exploit with traditional means [6]. In most cases, human beings have to transform context data into the input for business process management. In effect, large amounts of context data are not used because the effort to prepare it is too high [7]. The contribution of this paper is to prepare the methodological exploitation of context data by introducing a classification. There are two basic classes of context data, intrinsic and extrinsic context data. Intrinsic data has been created from within the
Final version available at http://link.springer.com/chapter/10.1007/978-3-319-15895-2_37
business process lifecycle (e.g. a comment on a process model). Extrinsic paper originate from outside, such as a blog entry by a customer. The paper is structured as follows. In the following section, a classification of business process context data into extrinsic and intrinsic data is introduced. Related work is analyzed in the following section. Finally, an outlook and the conclusion is given.
2
Classification of business process context data
Context data associated with business processes can be differentiated whether the data has been created extrinsically or intrinsically as shown in Fig. 1. Extrinsic data is created independently from the business process lifecycle [8]. Intrinsic data of a business process is created during the business process lifecycle. Context data
Extrinsic Data
Enterprise Architecture
Legal
Intrinsic Data
Social data
Design phase
Implementation phase
Deployment phase
Operation phase
Evaluation phase
Social production
Weak ties
Egalitarian decisions
Co-creation
Fig. 1. Classification of extrinsic and intrinsic context data
2.1
Extrinsic Data
Extrinsic data is differentiated into data describing enterprise architecture, legal documents and data created by social software, so-called social data [9] [10]. Enterprise architecture. Enterprise architecture [11] [12] is a static view on the enterprises defining the relationship of business units and IT-components. It also describes how to align business and IT in order to realize the company’s goals and implement the company’s strategy. Therefore, defining enterprise architecture is a crucial task of management. Enterprise architecture is connected with Taylorism [13] that means dividing up larger tasks into smaller, assigning them to individuals and controlling the execution of the tasks. Nevertheless a large amount of semi- and unstructured data associates these formal models (e.g. in TOGAF [14]).
Legal data. Legal data [15] shall be defined as data given by the lawmaker. It contains laws, regulatory status and implementation comments that influence the design of the business process and their operation. Social data. Enterprise 2.0 and social software replace the Taylorism [16] -oriented production of goods and provisioning of services by a bottom-up organized, egalitarian and cocreation oriented one. The four key concepts weak ties [17], social production [2] egalitarian decision-making [18] and co-creation [19] imply the creation of vast amounts of unstructured data. By analyzing these data, valuable information influencing one or more perspectives of business processes may be gathered. These perspectives are the organizational, operational, control, data and functional perspective. Weak ties [17] are informal relationships across the formal organization of enterprises and organizations. They are created to exchange knowledge and to combine competencies in order to handle tough problems. Weak ties can be identified by analyzing the competencies, the areas of work of the employees and by created artifacts. Especially artifacts may be helpful to discover hitherto unknown colleagues working on the same themes. By analyzing weak ties the organizational perspective of business process models can be enriched both by organizational relationship and competencies. Social production [2] inverses the product design of tayloristic approaches. Instead, a top-down-approach a bottom-up approach is used. The product is created by the number of individual contributions. Instead of realizing the master-plan of the topmanagement, the individual stakeholders may introduce their ideas and thoughts on the design of the product. Social production uses an inverse mechanism of quality control. Instead of pre-defining rules for measuring the quality of contributions it uses an a posteriori and holistic approach. Each contribution of an individual is visible to the public. Therefore, the individuals strive for a high quality in order to avoid damage to their social reputation. Social production provides information for the operational and functional perspective. Enterprise 2.0 and social software [1] also change the way decisions are made within enterprises. Instead of hierarchy based decisions, more egalitarian decisions are made. This implements concepts such as the wisdom of the crowds that prefer combining the opinions many stakeholders instead of relying on the decisions of a few specialists. The co-creation of products [20] is replacing the separation of consumer and producer and unidirectional, limited interaction by a bidirectional, more frequent interaction between a producer and prosumer. A prosumer [21] is a consumer actively participating in the design of a product. During the interaction of producer and prosumer, a lot of valuable information is exchanged to improve products and services. It may influence the operational control and functional perspective.
Organizational Weak Ties
Operational
Control
Data
Functional
+
Social production Egalitarian decision making Co-production
+
+ + +
Table 1. Social data and its impact on business process perspectives
2.2
Intrinsic data
Intrinsic data is created during the business process lifecycle [8]. It may be attached directly to a process element, e.g. an explanation of a task in a process graph, or it may refer to larger parts or the whole process or even process group. Often overlooked, but nevertheless of increasing importance are semi- and non-structured data associated with structured process definitions. Business process models using notations (e.g. BPMN [21]) and are often accompanied by non-structured data such as text files, in order to give further explanations. Text files contain explanations of business models. Furthermore comments of users contain suggestions for improving business processes and governance documents (like rules for the process design). During process design, interviews and questionnaires deliver both semi- and unstructured data. This data also embraces comments and suggestions for improving an existing process. In the implementation phase, documentation describing the relationship between the abstract process model and its implementation in the company's organization and software systems is created. Also, documents created during the deployment phase, that means the workflow to put the process into operation, may contain valuable data for other phase, e.g. the operation phase. In the operations phase, huge amounts of semi-structured data is collected by tracing the process execution in log files. These log files contain data about the history of the process executions consisting of execution time, duration, resource consumptions, etc. In the evaluation phase, evaluations and questionnaires are made in order to collect data how to improve the business process.
3
Related work
There are a number of approaches for exploiting context data. A first approach to automatically extract extrinsic information from policy documents is described in [22]. In [7] and [23] approaches for process models from / to natural language text is described. The extraction of workflow models from maintenance manuals is described in [24]. Process mining [25] uses semi-structured information from event-logs in order to discover processes, verify the compliance of process execution etc.
4
Conclusion and outlook
Data in the context of business processes is a valuable source of information for business process management. Context data may be extrinsic or intrinsic and is often unstructured. Advances in data processing allow to use context data to a far broader extent than before. Therefore, this paper showed how to leverage context data for business process management. Academics can improve from a new classification of context data from business processes and can so adopt and improve current approaches. For instance techniques of quality management of BPM and integration of social software can be improved. Practitioners can use e.g. the classification to evaluate and improve current BPM implementations. Future research can explore approaches for automatic context data analytics (e.g. with Text Mining methods) as well as industry sector specific adoptions of the classification and Big Data [26] based approaches.
References Andrew P. McAfee, “Enterprise 2.0: The Dawn of Emergent Collaboration,” MIT Sloan Manag. Rev., vol. 47, pp. 21–28, Apr. 2006. [2] Y. Benkler, The Wealth of Networks : How Social Production Transforms Markets and Freedom. Yale University Press, 2006. [3] F. W. Taylor, “The principles of scientific management,” N. Y., vol. 202, 1911. [4] W. Y. Chang, H. Abu-Amara, and J. F. Sanford, “Introduction to Enterprise Services and Cloud Resources1,” in Transforming Enterprise Cloud Services, Dordrecht: Springer Netherlands, 2011, pp. 1–42. [5] R. Blumberg and S. Atre, “The problem with unstructured data,” DM Rev., vol. 13, pp. 42–49, 2003. [6] J. Herbst and D. Karagiannis, “An inductive approach to the acquisition and adaptation of workflow models,” in Proceedings of the IJCAI, 1999, vol. 99, pp. 52–57. [7] F. Friedrich, J. Mendling, and F. Puhlmann, “Process model generation from natural language text,” in Advanced Information Systems Engineering, 2011, pp. 482–496. [8] M. Weske, Business Process Management: Concepts, Languages, Architectures. Springer-Verlag Berlin Heidelberg, 2007. [9] R. Schmidt, “Social Data for Product Innovation, Marketing and Customer Relations,” presented at the BPMS2, Tallinn, Estonia, 09-Mar-2012. [10] M. Brambilla, P. Fraternali, and C. Vaca, “BPMN and Design Patterns for Engineering Social BPM Solutions,” in Business Process Management Workshops, vol. 99, F. Daniel, K. Barkaoui, S. Dustdar, W. Aalst, J. Mylopoulos, M. Rosemann, M. J. Shaw, and C. Szyperski, Eds. Springer Berlin Heidelberg, 2012, pp. 219–230. [11] J. W. Ross, P. Weill, and D. Robertson, Enterprise Architecture as Strategy: Creating a Foundation for Business Execution. Harvard Business School Press, 2006. [1]
[12] R. Schmidt, M. Möhring, A. Zimmermann, M. Wissotzki, K. Sandkuhl, and D. Jugel, “Towards a Framework for Enterprise Architecture Analytics,” in Proceedings of the 18th IEEE International Enterprise Distributed Object Computing Conference Workshops (EDOCW), Ulm / Germany, in press., Ulm, Germany, 2014. [13] R. H. Guest and H. G. J. Aitken, “Taylorism at Watertown Arsenal: Scientific Management in Action 1908-1915,” Technol. Cult., vol. 2, no. 2, p. 191, 1961. [14] A. Josey, TOGAF Version 9 A Pocket Guide. Van Haren Pub, 2009. [15] G. Governatori, Z. Milosevic, and S. Sadiq, “Compliance checking between business processes and business contracts,” in Enterprise Distributed Object Computing Conference, 2006. EDOC ’06. 10th IEEE International, 2006, pp. 221–232. [16] F. W. Taylor, The Principles of Scientific Management. General Books LLC, 2010. [17] M. Granovetter, “The Strength of Weak Ties,” Am. J. Sociol., vol. 78, no. 6, pp. 1360–1380, 1973. [18] J. Surowiecki, The Wisdom of Crowds: : Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations. Anchor, 2005. [19] C. K. Prahalad and V. Ramaswamy, “Co-creating unique value with customers,” Strategy Leadersh., vol. 32, no. 3, pp. 4–9, 2004. [20] S. Vargo, P. Maglio, and M. Akaka, “On value and value co-creation: A service systems and service logic perspective,” Eur. Manag. J., vol. 26, no. 3, pp. 145– 152, Jun. 2008. [21] S. Klein and C. Totz, “Prosumers as service configurators-vision, status and future requirements1,” E-Life Dot Com Bust, p. 119, 2004. [22] J. Li, H. J. Wang, Z. Zhang, and J. L. Zhao, “A policy-based process mining framework: mining business policy texts for discovering process models,” Inf. Syst. E-Bus. Manag., vol. 8, no. 2, pp. 169–188, Mar. 2010. [23] H. Leopold, J. Mendling, and A. Polyvyanyy, “Generating natural language texts from business process models,” in Advanced Information Systems Engineering, 2012, pp. 64–79. [24] P. Schumacher, M. Minor, and E. Schulte-Zurhausen, “Extracting and enriching workflows from text,” in Information Reuse and Integration (IRI), 2013 IEEE 14th International Conference on, 2013, pp. 285–292. [25] W. M. Van der Aalst and A. Weijters, “Process mining: a research agenda,” Comput. Ind., vol. 53, no. 3, pp. 231–244, 2004. [26] R. Schmidt, M. Möhring, S. Maier, J. Pietsch, and R.-C. Härting, “Big Data as Strategic Enabler Insights from Central European Enterprises,” in In: Abramowicz, W., Kokkinaki, A. (Hrsg.). 17th International Conference on Business Information Systems (BIS). Springer LNBIP, vol. 176, Larnaca, Cyprus: Springer Berlin Heidelberg, 2014, pp. 50–60.
