Geração automática de modelos de processo em sistemas de Workflow

Abstract

Workflow systems have been widely employed; however, the automatic generation of process models, specially applied to the web service domains is still an area to be explored. Workflow problems are hard due to the size of the instances that generates a wide search space and requires a long time to be processed. Thus, this problem can be coped with the use of AI planning techniques associated with genetic algorithm. The first technique has many similarities with workflow and the second one is adequate for problems with a large search space. In this context, this work presents an architecture based on the use of a genetic planner in order to obtain the automatic generation of models for Workflow.