Exploring the Business Process Agility Issue: An Experience Report

Introduction

Business process agility (or flexibility) has been a matter of interest for numerous researchers (Milanovic et al., 2011; Van der Aalst et al., 2009; Snowdon et al., 2007; Pesic et al., 2007; Daoudi & Nurcan, 2007; Reijers, 2006; ShuiGuang et al., 2004; Rinderle et al., 2004; Mangan & Sadiq, 2002; Millie & Balasubramanian, 1997). Agility in the context of business processes can be defined as the ability of an organization to effect changes in the process components (activities, roles, resources, information etc.) in a timely manner, usually in response to changes in business environment and stakeholders’ needs (Alexopoulou et al., 2008). The intense interest in business process agility stems from the fact that business process automation supported by the utilization of process-aware information systems (Dumas et al., 2005) has increased accuracy and efficiency in process execution on one hand, but it has also rendered business process modification a complex and time-consuming task. This is because well-structured business process models executed by Business Process Management Systems (BPMS) (Dumas et al., 2005) proved to be inflexible. Since modern enterprises operate in highly turbulent environments having to cope with a frenetic pace of change (van Oosterhout et al., 2006) and continuously sense opportunities for competitive action in their product-market spaces, it is business process agility, which underlies enterprises’ success in constantly enhancing and redefining their value creation in highly dynamic environments (Sambamurthy et al., 2003).

In an effort to realize business process agility, researchers propose various methods, techniques or approaches in general, focusing on business process automation. One paradigm for the creation of agile business processes is based on the definition of abstract or loose models (Lin & Orlowska, 2005; Herrmann et al., 2000) as opposed to strict, well-structured ones. Van der Aalst (1999), for example, proposes a general model from which several variations may be produced during execution through inheritance, which constitute specific models. ShuiGuang et al. (2004) suggest a modeling method according to which a business process model includes, apart from concrete parts of activity flows, some unspecified parts whose structure is formed during execution time, based on current conditions and by picking the appropriate activity from a ‘pool’ of activities. The same idea lies behind ‘flexibility pockets’ (Sadiq et al., 2001) that represent activities that are orchestrated into specific flows during execution time. Other researchers contribute in the solution of the business process agility issue through the formal definition of modification rules (Casati et al., 1996; Weske, 2001; Ellis et al., 1995). ADEPT_flex (Dadam & Reichert, 2009) is an endeavour based on this logic. It represents a set of operations, which enable the modification of business processes under execution, ensuring that structural correctness and consistency of the running instances is retained. Other approaches adopt formalisms based on the specification of constraints (Sadiq et al., 2005). Such an approach, for example, is proposed by Dourish et al. (1996), according to which instead of following a predetermined order of activities for the execution of a business process, actors are free to choose the activity they want to execute, provided that they do not violate specific constraints.