Mixing Planning and Scheduling to Model Complex Process Environments

In most current APS (Advanced Planning and Scheduling) systems the planning and scheduling tasks are handled separately using different methods and technology from the areas like Artificial Intelligence and Operations Research. Recently, Constraint Programming becomes a roof over several solving technologies that allow us to solve both planning and scheduling tasks within single declarative framework. In the paper, we suggest mixing planning and scheduling tasks within single system that is capable to solve more complicated scheduling problems in complex process environments. We analyse different views of planning and scheduling, identify the similarities and the differences of both tasks and we propose a general structure of scheduler with some planning capabilities. In the second part of the paper, we compare various modelling approaches from the mixed planning and scheduling point of view. We concentrate on capabilities of the models to capture typical problems in complex process environments primarily but the conclusions are applicable to other (non-process) problem areas. The presented results make the basics of a generic scheduling engine that is currently implemented within the VisOpt project.