Skip to Main Content
In the last years, autonomous and adaptive processes have became more and more important. Even business processes are more complex today. The more complex a system becomes, the more difficult it is to handle changed requirements. This increases the need for adaptation abilities of workflows. In these new workflow systems, autonomy would have great impact regarding the possible usability of such workflows. They can decide on their own how to handle a certain situation, for example, in case of a Web service timeout, new business rules, or new knowledge about the domain. Another aspect in current workflow engines is the inefficient way of executing process instances. Today, process instances perform actions separated from each other. This leads to much redundant work if two process instances work on the same workflow model at the same time. Some of the tasks will be executed twice, but the result in every process instance is the same.Without knowing what all the other instances do, the engines waste time and resources. Our primary goal is to provide an approach for autonomous adaptations in workflow engines. We solved that through a combination of the Belief-Desire-Intention technique and semantics. We also provide processes with the ability to know all other engines and processes running and to communicate with other engines as well as with all processes via semantic knowledge.