REPLAY - Replaying History on Process Models for Conformance Checking and Performance Analysis

The availability of process models and event logs is steadily increasing as more and more business processes are supported by IT. On the one hand, most organizations make substantial efforts to document their processes, while on the other hand, these processes leave footprints in their information systems. Although it is possible to extract event logs from today's systems, the relation between process models and event logs is seldom investigated. Yet, the availability of event logs on the one hand and models on the other hand enables conformance checking, i.e., investigating whether reality deviates from a priori defined models. This is useful for a variety of reasons, e.g., to show compliance or to improve process support. Moreover, the logs can be used to extend and repair process models. In particular, the process models may be automatically augmented with performance information showing, for example, bottlenecks. One can go even one step further and use historic information to predict the performance or compliance of running cases.

The focus of the REPLAY project is on conformance checking and performance analysis by replaying history on the process model. In case of substantial deviations, logs and models of different granularity, and advanced modeling constructs, this is far from trivial; it is comparable to following a trail where parts of the track are missing. The REPLAY project addresses such problems and subsequently applies the results in three application domains: hospital information systems, municipal information systems, and high-tech deployed systems. The results will be implemented and made available through our open-source process mining framework ProM. The project is funded by NWO.