A Complementary Analysis of the Behavior of BPMN Tools Regarding Process Modeling Problems.

Business process models are essential for organizations, enabling participants to understand the business processes in which they are involved. These models are mainly designed using process modeling tools, supporting Business Process Model and Notation (BPMN) 2.0, which is widely accepted by the community. However, representation of modeling problems in process models may generate inconsistent interpretations, leading to the implementation of incorrect modeling solutions. As such, BPMN 2.0-based process modeling tools should detect these problems. The literature shows that modeling tools behave differently when facing identical problems. This paper analyzes how BPMN 2.0-based process modeling tools currently react and provide feedback about modeling problems in business process models. Process modeling anti-patterns are used as study cases; they compose a class of commonly recreated modeling bad practices. This paper also reviews and complements experiments from the literature to understand the current state of problem detection by modeling tools. Each of the ten anti-patterns is modeled in ten modeling tools. An analysis of which types of problems are more often detected and how modeling tools react to them is presented. Furthermore, problem feedback should be displayed understandably. So, problematic models are created, and visual feedback about their problems is generated according to recommendations from the literature. These problematic models are then introduced into the modeling tools. The tools' reactions are compared to the literature recommendations to evaluate the current gaps in visual feedback presented by modeling tools.