Automatic Support for Product Based Workflow Design: Generation of Process Models from a Product Data Model

Abstract

Abstract. Product Based Workflow Design (PBWD) is one of the few scientific methodologies for the (re)design of workflow processes. It is based on an analysis of the product that is produced in the workflow process and derives a process model from the product structure. Until now this derivation has been a manual task and is therefore a time-consuming and error-prone exercise. Automatic support would enhance the use of the PBWD methodology. In this paper we propose several algorithms to automatically generate process models from a product structure and we present a software tool (implemented in ProM) to support this. Finally, the properties of the resulting process models are analysed and discussed.

Product-Based Workflow Design for Monitoring of Collaborative Business Processes

“…4). Several algorithms have been proposed to transform the PDM into a process model executable, for instance, by a workflow management system [22]. We use the algorithm described in [19] to automatically generate a process model for the optimal path of the PDM.…”

“…The design of the monitoring process for cross-organizational business processes is framed as a PBWD (Product-Based Workflow Design, [16,20,22]) problem. PBWD is an analytical method for automatically deriving business process specifications from the set of information products involved in the process and their dependencies.…”

A stage-based model for enabling decision support in process mining

“…The second validation situation is to execute the dependency equations without loops equations using FEL 1 is supported by other concepts, namely, log-based ordering relations [103], the abstract representation to discover a model [106], the bottom-up strategy [107] and other dependency equations [58,115]. …”

“…Thus, the spaghetti-like model can be lasagna-like and provide better support to decision-making. To achieve this contribution, the proposed method is consecutively executed by creating new algorithms and matrices, since composing the model manually is time-consuming [107]. When composing the model with stages, this sequential execution considers the typical challenges, namely, loops and parallelism, which should be addressed in any effective process mining technique [92].…”

“…Vanderfeesten et al [107] proposed an algorithm called Foxtrot that uses the bottom-up strategy to generate a process model from a product data model. However, Foxtrot can not handle parallel execution.…”