Goal-driven multi-process analysis

Extant process modeling techniques address different aspects of processes, such as activity sequencing, resource allocation, and organizational responsibilities. These techniques are usually based on graphic notation and are driven by practice rather than by theoretical foundations. The lack of theoretical principles hinders the ability to ascertain the "correctness" of a process model. A few techniques (notably Petri Nets) are formalized and apply verification mechanisms (mostly for activity sequencing and concurrency). However, these techniques do not deal with important aspects of process design such as process goals. As previously suggested, a formal process modeling framework, termed the Generic Process Model (GPM), has been used to define the notion of process model validity. In GPM, validity is based on the idea that the purpose of process design is to assure that an enacted process can reach its goal. In practice, often several processes "work" together to accomplish goals in an organizational domain. Accordingly, in this paper we extend the validity analysis of a single process to a "cluster" of processes related by the exchange of physical entities or information. We develop validity criteria and demonstrate their application to models taken from the Supply Chain Operations Reference-model (SCOR). We also use the formal concepts to analyze the role of an information system in inter-process communication and its possible effects on process cluster validity.