Towards a Probabilistic Approach for Better Change Management in BPM Systems

The business process modeling has become a mature subject in the research of information systems domain. Although, the business process models remain in consistent evolution to be available for the enterprises to better respond to the market needs and to inflect the needs of different stakeholders, in consequence to construct and preserve their competitive advantages. In this regard, we propose a declarative (rule-based) approach which exploit the relationships among the artifacts (e.g. activities, data, and roles) of a business process model. It may also consider the context of the change and the history of dependencies among different elements of the process. We may then experiment the change impact prediction using inference made by either the probabilities, knowledge-base, or evolutionary algorithms. It may help to better assess the volume of impact, as a result of the planned change.