This project is developing methods and tools that allow people to define robust workflow models much more quickly and effectively. This work is led by Mark Klein and Chris Dellarocas.
Today's business environments are characterized by uncertain requirements and rapid and continuous change. In order to effectively support business processes in such contexts, workflow systems must be able to respond effectively when deviations from the "ideal" process occur during process execution. Such "exceptions" can include process enactment errors, violations of the assumptions (e.g. concerning resource availability) underlying the current process model, or even changes in the business environment not yet reflected in the current process model. If not detected promptly and handled effectively, such exceptions can result in severe impacts on the effectiveness of collaborative work.
Workflow systems currently provide little support for exception handling. Most do not allow one to modify a process model once it has started executing. Exceptions are handled by attempting to include conditional branches for all possible contingencies. It is difficult, however, to anticipate all possible failures. Adding such branches also greatly complicates the process models and thereby obscures the "preferred" process. Even the systems that do support dynamic workflow model modification do not help determine the best response to a given exception.
This project is developing a knowledge-based approach to handling exceptions in process enactment systems. Rather than requiring process definers to anticipate all possible exceptions based on their intuitions and experience, this approach uses an extensible collection of generic exception handling knowledge that suggests what kinds of exceptions to check for, how they can be best detected, and how they should be resolved. While this project is focused on workflow model definition, the same methodology and knowledge base can be used when designing business processes that are not enacted [in their entirety] by workflow systems.
The following white paper contains more information about this project.
Related Publications