Towards self-adaptation and evolution in business process

Business process run-time evolution and adaptivity are two urgent objectives in the research agenda of dynamic workflow execution. Traditional languages as BPMN or BPEL take an imperative style for defining the exact sequences of activities to execute. The imperative approach identifies a narrow space of solution that is generally optimized by the experience. However it does not provide enough freedom of action to bypass obstacles when something exceptional happens. In the wake of declarative specification languages we propose a framework, based on the standard BPMN, in which both business goals and the operative context are monitored for changes during the execution time. To enable a flexible adaptivity of the process to a changing environment we adopt the solution of relax some constraints of the rigid BPMN specification thus to give autonomous software agents the opportunity of exploring a wider space of solution, even when this space evolves unexpectedly or contains uncertainty. The result is a multi-agent system that exploits its features (mainly autonomy and proactivity) in order to monitor the execution state of the process and to discover a distributed solution to unpredictable situations or to specifications' evolution.