|
Abstract : |
Workflow management systems (WFMS) can be used in CE to manage the increasing communication and coordination efforts. For this purpose WFMS require an explicit model of the processes that are to be supported. Creating such a workflow model and adapting it to changing requirements is a time consuming and error prone task. For this reason we are investigating induction techniques from machine learning to support these tasks. Our previous approach (Herbst, 1999) to workflow induction was restricted to sequential processes. In this contribution we show, how the basic idea of our previous approach can be generalized, so that it is able to deal with concurrent processes. The applicability is demonstrated using a simplified version of a release process of the Mercedes Benz passenger car division., |
