Towards Situational Business Process Meta-Modelling Oumaima Saidani1, Selmin Nurcan 1,2 1 Université Paris 1 - Panthéon - Sorbonne Centre de Recherche en Informatique 90, rue de Tolbiac 75013 Paris, France, 2 IAE de Paris Sorbonne Graduate Business School Université Paris 1 - Panthéon - Sorbonne 21, rue Broca 75005 Paris France {Oumaima.Saidani, Selmin.Nurcan}@univ-paris1.fr Abstract. Business Process (BP) meta-models allow partial views of the processes. There may be adequate for some processes but not others. Situational engineering has proved its effectiveness in many engineering domains such as software and information system development. Reasoning on a situational approach for BP meta-modelling is a challenging research issue which can contribute to increase flexibility of meta-models and their adaptability to different organisation settings. Keywords: Business Process meta-modelling, Flexibility, Adaptability. 1 Introduction Current researches on business process (BP) modelling stress the importance of the flexibility and the adaptability of BP [2], [5] [7]. Reasoning on variability in modelling artifacts can meet the flexibility and context-awareness requirements by offering alternative solutions depending on the context and on the point-of-views of the decision-makers. A BP model is often formalized, at the type level, using a meta- model which captures the concepts supported by this model. We promote the idea that a single BP meta-model is still insufficient. A promising idea is to propose an approach for adapting and configuring existing meta-models according the organisation settings and users’ objectives, rather than to advice for a single model which can be too complex for some requirements and simple for others. Accordingly, we focus on the flexibility at the type (meta-model) level of the BP which corresponds to the level 2 of the OMG four-level-architecture for the processes [1]. 94 Proceedings of CAiSE’08 Forum BPs are of various kinds and are defined in different levels of abstraction using various artifacts depending on the organisation settings and the purpose of the modelling. For instance, in mechanistic or production organisations, they are often prescribed in a detailed level since they shall be executed. On contrary, in adhocracies organisation, more freedom can be left to business actors for choosing how to perform the underlying business objectives. Therefore, the meta-models can be different and capture only some aspects of processes, however, sometimes their interrelationships could or should be taken into consideration and their complimentarily needs to be expressed. That is, in some situations, activity-oriented and product-oriented ones may need to be matched in order to determine which activity influences on which product and on which moment of the process. Also, strategy-oriented process meta- models require to be made operational using activity-oriented meta-models [3]. As well, [8] combines intention-oriented and state-based process modelling. Therefore, mechanisms for adapting existing models to specific requirements need to be developed. Our aim in this paper is to propose such mechanisms. Our motivation behind this proposal is that: (i) a BP meta-model which is designed for a specific organisation setting is not necessarily adequate for others; (ii) since several meta- models have proved their effectiveness in many business areas, it does not seem required to create new models. In the information systems development (ISD) community, method engineering (ME) has been introduced as a response to the need for methods adapted to specific ISD project situations, and to the failure of the methods known as "universal" [9]. One area of ME is Situational Method Engineering (SME). SME is based on four principles: meta-modelling, flexibility, reuse and modularity [10]. We can highlight that the ISD requirements on flexibility and adaptability that are behind the ME emergence in the ISD field were similar to those currently observed in the BPM field, we thus base our reasoning on SME mechanisms. The paper is structured as follows. Section 2 introduces an overview of the proposed approach with illustrative examples. Section 3 concludes the paper. 2 Overview of the proposed approach with examples Building the adequate meta-model can be done following several manners, for instance, by assembling relevant concepts, by constructing a core meta-model and enhancing it with required concepts, etc. With analogy to the method in the ISD field, we introduce the concept of business method which consists of a set of reusable components that we identify as BP meta-model chunks. In the remainder of the paper, we simply denote them by BPM-chunk. BPM-chunks are independent and stored in a chunk repository. They can be reused in order to build new meta-models or to enhance existing ones. The can be simple (e.g. a concept) or compound (e.g. a set of concepts, properties and relationships between them). In the reminder, we introduce some examples of BPM-chunks that constitute a partial vision of the repository. We underline the use of some operators for managing them. We are inspired from operators defined in [4]. Fig. 1 shows an example of meta-model (M0) which can be Proceedings of CAiSE’08 Forum 95 extended, according to the situation, by independent chunks (C0, C1, C2) resulting on the meta-models shown in Fig. 1 (right). PM0 and PM1. PM0 (Fig. 1 (left)) keeps a minimal set of features. It may be suitable for some organisation settings, e.g. stable organisations with minor changes and few operations. Otherwise, defining operations in a finer granularity, and in frequently changing organisations, may involve a cumbersome work. In such situation, PM0 can be extended with C0 (Fig. 2) in order to construct PM1. C0 serves, in PM1, as a link between roles and operations, BPs are relied to functions rather than operations. PM1 is discussed in detail in [5]. Extending PM0 requires updating the relationships can- hold and comprises and defining a new one: satisfies. Let CONCEPTS the set of concepts of the chunk repository. A relationship can associate many concepts. Formula (1) represents the mapping of a relationship r onto a set of concepts. Let create-relationships, update-relationships and delete-relationships three operators allowing respectively creating, updating and deleting relationships between entities. These operators can be applied so that the relationship can-hold between the entities Role and Operation -in PM0- is removed, and the same is created in PM1. As well the relationship Comprises between the entities Business-Process and Operation, in PM0, are removed and those between Business-Process and Function are created in PM1. Finally, the relationship Satisfies between Operational-Goal and Operation is created. relationship − concept(r : RS ) → 2 CONCEPTS , relationship − concept (ri ) ⊆ CONCEPTS (1) CCxPM1 Organizational Function and operational goal extension Unit * Belongs to CCxPM0 * Can play Can hold CxPM1 CPM1 Actor Role Operation * * * 1 * * * * Participates CxPM0 CPM0 Business PM1 Process * Comprises * Acts on C n Reachs * on io ns te PM0 1 Business xt te ex ex Business goal object te t in ns t ra io n ns Co Fig.1 The Meta-model of PM0 (left) and a set of BP meta-models and their relationships (right) Comprises (C0) Function Operational goal (C1) Context (C2) Constraint 1 * Fig.2 Examples of BPM-chunks representing respectively C0, C1 and C2. CxPM. CxPM0 and CxPM1 extend PM0 and PM1 with chunk C1 (Context) (Fig. 2). C1 can be added to an existing meta-model for capturing context knowledge which can impact the assignments relationships of a process model (e.g. the ability of actors for playing roles according to a given context [6]. CxPM0 is constructed by extending 96 Proceedings of CAiSE’08 Forum PM0 with C1. The integration of PM0 and C1 requires the use of the operator update- relationships so that the relationships can-play, can-hold, implies and comprises defined in PM0 are related also to C1. The same logic can be applied for CxPM1. CPM. In some situations, organisation policies need to be enforced impacting assignments decisions, for instance, separation of duties (see [5] for more details about constraints). Building CPM0 (resp. CPM1) requires extending PM0 (resp. PM1) with C2 (Fig. 2). This practice needs using the operator update-relationships so that the constrained binary relationships assignment (e.g., can-play) in PM0 (resp. in PM1) are related to C2 in CPM0 (resp. CPM1). The same reasoning can be applied for CxPM1. 4 Conclusion and Future Work This paper provides a start points for the definition of a methodology allowing the design of adaptive and flexible BP meta-models according to the situation at hand. We have introduced the concepts of BPM-chunk and business method as well as example of chunks and meta-models in order to illustrate our proposal. We promote the fact that the final business process model has to be created from the set of proposed chunks in order to suit to a particular situation. This approach aims to make easier the definition of flexible and customised meta-models. Dealing with situation- awareness raises many issues which need further research such as: the context influencing the selection of adequate chunks and the adaptation process. References 1. BPMI.org, OMG: Business Process Modelling Notation Specification. Final Adopted Specification. Object Management Group (2006) 2. Nurcan, S.: A survey on the flexibility requirements related to business processes and modelling artifacts. Proc. HICSS’08, Big Island, Hawaii, USA, 7-10 January (2008) 3. Nurcan, S., Etien, A., Kaabi, R., Zoukar, I. and Rolland, C.: A strategy driven Business Process Modelling Approach. Business Process Management Journal, Vol. 11, N° 6 (2005) 4. Ralyté, J.: Ingéniérie des méthodes à base de composants. PhD thesis, University of Paris 1- Sorbonne (2001) 5. Saidani, O. and Nurcan, S.: A Role-Based Approach for Modelling Flexible Business Processes. In proceedings of BPMDS’06 (2006) 6. Saidani, O., Nurcan, N.: Towards Context Aware Business Process Modelling, BPMDS’07, Held in conjunction with CAiSE’07, Trondheim, Norway (2007) 7. Snowdon, R. A., Warboys, R. M. Greenwood, C. P. Holland, P. J. Kawalek, D. R.: On the architecture and form of flexible process support. SPIP Journal, Vol. 12, N° 1 (2007) 8. Soffer, P., Rolland, R.: Combining Intention-Oriented and State-Based Process Modelling. ER (2005) p. 47-62 9. Rolland, C.: L’ingénierie des méthodes : une visite guidée, e-TI - la revue électronique des technologies d'information, Premier Numéro, 25 octobre 2005 (2005) 10. Rolland, C.: Method Engineering: Trends and Challenges. In IFIP International Federation for Information Processing, Vol. 244, Situational Method Engineering: Fundamentals and Experiences, eds, Ralyté, J., Brinkemper, S., Henderson-Sellers, B., (Boston Springer) (2007)