Management and utilization of knowledge for the automatic improvement of workflow performance

We present a framework that enables reengineers to build a base of performance improvement knowledge that can be used to automatically improve workflow performance. Automatic improvement of workflow performance involves modification of a business information system such that the predicted performance of its business workflows satisfies a performance goal. The number of possible modification options is very large, so a significant body of knowledge is needed to choose among them. We demonstrate, using a simple example, the requirements for the types of knowledge necessary in a automatic improvement framework. We define a knowledge model for representing these types of knowledge. We use the model to provide the framework with a body of domain-independent performance improvement knowledge. We then describe how the framework enables reengineers to provide additional performance improvement knowledge to the model and how the framework utilizes that knowledge to automatically improve workflow performance to meet the performance goal.