This research will study the problem of scheduling a complex job shop system under dynamic arrivals. This will include product routings that may be reentrant flows. Machines may incur sequence- dependent setup times. This is an important, but difficult problem to solve. The research will provide analytical models as well as heuristics for the simpler versions of the multi-machine problem and, in particular, the bottleneck machines. These analytical models will be used to provide a decision support system for shop flow scheduling. The ultimate goal is to develop efficient algorithms for the more complicated and realistic problems by adopting the techniques and properties developed for the simpler problem in a more sophisticated or "smart and lucky" search algorithm. This problem is inspired by actual industrial problems and will closely collaborate with the personnel of a manufacturing job shop to test the algorithms and systems in an industrial setting.
