Workflow management systems, business process management systems, enterprise information systems, and e-business systems are increasingly constructed using generic software packages. The technology to generate or assemble process-aware information systems is available but still limited. Therefore, the evaluation of existing software and languages and the development of prototypes to improve current practice are highly relevant. An example of a project would be to extend a workflow engine (e.g., FLOWer or IBM’s WebSphere) with advanced case prediction facilities (e.g., When will a particular customer order be handled?).

In recent years, process mining has emerged as a way to analyze systems and their actual use based on the event logs they produce. Information systems, but also all kinds of devices, increasingly generate audit trails. This information can be used to automatically generate models allowing analysis. Unlike classical analysis techniques that focus on some subjective model, process mining takes reality as a starting point. The ProM tool can e.g. reveal conceptual errors of practitioners about the processes they are involved in. An example of a Master project would be the development of process mining techniques for Philips Medical Systems aimed at finding patterns in the use of X-ray machines.

In practice a wide variety of languages are used. There are hundreds of workflow languages and also software development tools and simulation tools tend to use propriety languages or dialects of standardized languages. Moreover, these languages are typically not suited for automated analysis. Hence, there are many situations where models in one language need to be translated to some other language to allow for interoperability, enactment, or verification. This is particularly tricky for process models and thus it is challenging to find smart ways to translate these models for particular purposes. An example of an assignment would be to translate BPMN models into BPEL models for enactment in IBM’s middleware.

The use of formal methods enables all kinds of analysis including verification. Many organizations have models that need to be verified. For example, a recent study showed that more than 20 percent of the 600 process models in the SAP reference model have errors. Using Petri-net-based analysis techniques such errors can be detected, reported, and even predicted. An example of an assignment would be to translate the BPMN models of a large auditing firm into Petri nets and to try and detect deadlocks, livelocks, and other anomalies.

Having explicit process models typically enables the use of simulation techniques. Simulation can be used for validation and performance analysis. The architecture of a complex enterprise information system can be validated through interactive simulation. The expected performance of a high-end copier can be calculated through simulation. An example of a Master project could be the improvement of the simulation facilities of Protos using historic information coming from particular workflow systems.