Life-Cycle Support for Staff Assignment Rules in Process-Aware Information Systems

Process mining has been proposed as a tool for analyzing
business processes based on events logs. Today, most information systems are logging events in some log and thus provide detailed information about the processes they are supporting. This information can be used for two forms of process mining: conformance checking (comparing the actual process with some a-priori model) and discovery (deriving a
model from scratch). Most of the process mining tools have been focusing on the control-flow perspective and today it is possible to automatically construct process models that can be used for the configuration of Process-Aware Information Systems (PAISs). This paper provides an
overview of process mining and focuses on a neglected aspect of PAISs: staff assignment. We propose an approach for staff assignment mining based on decision tree learning, i.e., based on some organizational model and an event log we try to discover allocation rules. This is useful for configuring new PAISs. However, it can also be used to evaluate staff
assignment rules in some existing PAIS. Based on this, flaws and redundancies within staff assignment rules (e.g., security holes by offering process activities to non-authorized users in exceptional cases) can be
detected and optimization strategies can be derived automatically. The approach has been implemented in the context of the ProM framework and different strategies have been evaluated using simulation. Altogether, this work contributes to a complete life-cycle support for staff assignment rules.