A Learning Metaheuristic for the Multi Mode Resource Constrained Project Scheduling Problem

This abstract introduces a novel approach to intelligently select appropriate modes for each activity in the multi mode resource constrained project scheduling problem. The mode of an activity determines its duration and the resources that it requires. In order to obtain a good (or optimal) makespan, finding an appropriate mode for each activity is most important. The approach integrates different components. It applies a serial forward/backward heuristic to construct schedules; learning automata to come up with good quality modes and a genetic algorithm to explore the search space.