On Parallel Machine Scheduling Problem with Time Windows

Parallel Machine Scheduling Problem with Time Windows(PMSPTW) can be described as scheduling N jobs on M nonhomogeneous parallel machines in order to maximize the contributions of jobs scheduled. For a given job, it can only be processed on a subset of machines within several given time windows, and besides, the process time and time windows are machine dependent. To solve the problem, we introduce a Constraint Programming model for PMSPTW, and a system for integrating Constraint Programming and tabu search techniques. We only use the CP system to check the validity of solutions and determine the values of constrained variables, not to search for solutions. The search is performed by a tabu search procedure to avoid being trapped in local minima. In the paper end, experimental results for random generated benchmark test problems are also presented and analyzed.