BENCHMARKING REAL-WORLD JOB-SHOP SCHEDULING USING A GENETIC ALGORITHM WITH A SIMULATION APPROACH

Although production scheduling has attracted the research interest of production economics communities for decades, a gap still remains between academic examples and real-world problems. Genetic Algorithms (GA) constitute techniques which have already been applied to a variety of combinatorial problems. I intend to explain the application of a GA approach to bridge this gap for job-shop scheduling problems, by minimizing the makespan of a production program or increasing the due-date reliability of jobs. Simulation is a useful tool in problem solving. Here repetitive runs of simulated models or computed solutions through algorithms are applied. For job-shop and resource-constrained project scheduling, problems trying to bridge the gap between computed solutions and the feasibility of the simulation occur. I would like to explain the application of this special GA for job shop and resource-constrained project scheduling. Possible goals for scheduling problems include minimizing the time required of a production program, or increasing the due-date reliability of jobs, or possibly other objectives which can be described in a mathematical expression. The approach focuses on integrating a GA into a commercial software product, namely Microsoft Project 2003, and verifying the results with the simulation.