Learning, planning, and the life cycle of workflow management

This paper describes an approach towards workflow management based on the combination of learning and planning. Assuming that processes cannot be fully described at build-time, the approach makes use of learning techniques, namely inductive logic programming (ILP), in order to discover workflow activities as planning operators. These operators are subsequently fed to a partial-order planner in order to find the process model as a planning solution. The continuous interplay between learning, planning and execution aims at arriving at a feasible plan by successive refinement of the operators. The approach is illustrated in two simple scenarios. The paper concludes by relating the proposed approach with previous developments in this area.