Ensuring compliance with semantic constraints in process adaptation with rule-based event processing

An Adaptive Process Management System (APMS) allows for flexible, dynamic and even ad hoc adaptation of business processes based on case data, context and events. It is also important that APMS technology ensure error-free process execution and compliance with semantic constraints. However, most process design tools tend to be rigid or they handle only syntactic constraints. This restricts their value in real-world applications considerably. This paper presents a new approach to validate process change operations against semantic constraints using an integer programming formulation. The formulation allows us to describe existential as well as coordination (such as before-after ordering sequence) relationships between tasks in a process in a common way. It can then be solved to not only check full or strong compliance, but also determine the minimum set of additional process changes required to ensure weak compliance. Notions of strong and weak compliance are discussed and illustrated with a detailed example. We argue that this approach is more elegant and superior to a pure logic based approach.