One might naively think that any combination of a modern business process modeling language and a modern business rule modeling language would be able to fully model the dynamic and static structure of today’s enterprises. This study, based on an analysis of four process modeling specifications--colored Petri nets (CPNs), event-driven process chains (EPCs), the integrated definition for process description capture method (IDEF3), and the business process modeling notation (BPMN)--and four rule modeling specifications--the Simple Rule Markup Language (SRML), the semantics of business vocabulary and rules (SBVR), the semantic Web rules language (SWRL), and the production rule representation (PRR)--proves otherwise.

The authors use the Bunge-Wand-Weber (BWW) scheme (or ontology) to demonstrate that even the best pair of process and rule modeling languages--BPMN and SRML--only represents 23 of the 29 BWW constructs, with a noticeable overlap of six elements (20 percent) occurring in both languages.

Even though the basic idea and method of comparing the power of such combinations of languages is highly appealing, there are some shortcomings and unresolved questions in the paper that definitely warrant further exploration before one can use the results uncritically.

For instance, the authors claim that an EPC does not include constructs for representing systems and system decomposition. While this is true for the process-only part of an EPC, the currently used extended EPC does offer full system modeling capabilities. The paper also claims that an EPC is “not designed on the basis of a formal meta-model,” which is not true [1]. On the other hand, BPMN is said to possess such features, although I cannot find this in the specification (incidentally, the paper compares version 1.0, which has since been made obsolete by version 2.0).

This easy-to-follow paper includes useful definitions and introductions to the domains investigated and the methodology used. It should appeal not only to those in academia, but also to business process and rule modeling professionals who wish to obtain a deeper understanding of the shortcomings of current modeling approaches.