Workflow modeling languages allow for the specification of executable business processes. They, however, typically do not provide any guidance for the adaptation of workflow models, i.e. they do not offer any methods or tools explaining and highlighting which adaptations of the models are feasible and which are not. Therefore, an approach to identify so-called configurable elements of a workflow modeling language and to add configuration opportunities to workflow models is presented in this paper. Configurable elements are the elements of a workflow model that can be modified such that the behavior represented by the model is restricted. More precisely, a configurable element can be either set to enabled, to blocked, or to hidden. To ensure that such configurations lead only to desirable models, our approach allows for imposing so-called requirements on the model's configuration. They have to be fulfilled by any configuration, and limit therefore the freedom of configuration choices. The identification of configurable elements within the workflow modeling language of YAWL and the derivation of the new "configurable YAWL" language provide a concrete example for a rather generic approach. A transformation of configured models into lawful YAWL models demonstrates its applicability.

Abstract. A configurable process model captures a family of related process models in a single artifact. Such models are intended to be configured to fit the requirements of specific organizations or projects, leading to individualized process models that are subsequently used for domain analysis or solution design. This article proposes a formal foundation for individualizing configurable process models incrementally, while preserving correctness, both with respect to syntax and behavioral semantics. Specifically, assuming the configurable process model is behaviorally sound, the individualized process models are guaranteed to be sound. The theory is first developed in the context of Petri nets and then extended to a process modeling notation widely used in practice, namely Event-driven Process Chains.

No abstract

Configurable process models integrate different variants of a business process into a single model. Through configuration users of such models can then combine the variants to derive a process model optimally fitting their individual needs. While techniques for such models were suggested in previous research, this paper presents a case study in which these techniques were extensively tested on a real-world scenario. We gathered information from four Dutch municipalities on registration processes executed on a daily basis. For each process we identified variations among municipalities and integrated them into a single, configurable process model, which can be executed in the YAWL workflow environment. We then evaluated the approach through interviews with organizations that support municipalities in organizing and executing their processes. The paper reports on both the feedback of the interviewed partners and our own observations during the model creation.

Abstract. Reference process models capture recurrent business operations in a given domain such as procurement or logistics. These models are intended to be configured to fit the requirements of specific organizations or projects, leading to individualized process models that are subsequently used for domain analysis or solution design. Although the advantages of reusing reference process models compared to designing process models from scratch are widely accepted, the methods employed to configure reference process models are manual and error-prone. In particular, analysts are left with the burden of ensuring the correctness of the individualized process models and to manually fix errors. This paper proposes a foundation for configuring reference process models incrementally and in a way that ensures the correctness of the individualized process models, both with respect to syntax and behavioral semantics. Specifically, assuming the reference process model is behaviorally sound, the individualized process models are guaranteed to be sound.