Modelling and Managing Variability in Business Process Models

Abstract

Business Process (BP) models capture the coordination of a set of activities whose execution realizes specific business goals within a company. However, the construction of such models entails a big challenge for modellers and strongly depends on the nature of the domain being modelled. Moreover, when this nature involves handling many process alternatives, the use of variability modelling mechanisms becomes essential to succeed in the BP modelling task. Even though BP modelling variability has already been addressed by researches from the BPM community, it still remains as a challenge that is requested as hot topic in the most relevant conferences related to the BP area (i.e. the international conference on Business Process Modeling (BPM), CoopIS or BPDSM). This demand appears since the solutions provided in the literature do not deal with variability in a broad sense (considering all types of variability that we can find in a BP model), and in a scalable manner. In this context, this work provides a modelling approach that brings variability concepts as first order aspects of the modelling process. Concretely, the approach isolates the variability factors that affect a BP and allows managing independently their impact over the whole model. For such purpose, we rely on the techniques proposed in the field of the Software Product Lines to deal with variability issues. These techniques allows enhancing variability expressiveness as well as promoting model maintenance, legibility, understanding and reuse regarding variability. Finally, a running example is described and developed to illustrate the proposal and its applicability.