Lot streaming is the process of splitting a job or lot to allow overlapping between successive operations in a multistage system. The use of transfer lots usually results in a shorter makespan and lower work-in-process inventory than without it. In the thesis, we first study the discrete version of schedules which minimize the makespan for a single job in two- and three- machine flow shops. The optimal solution on s sublots can be obtained in O(s) time for the two-machine problem. We also present the first polynomial time solution for the three-machine problem. Two very good quality approximations for the discrete lot streaming solution in m-machine flow shops are also presented. Other topics studied are the effect of setups (either detached or attached) on continuous lot streaming for three-machine flow shops. We show that, in most cases, the optimal schedule and its structure change substantially in comparison with the no setup case. In the presence of either detached or attached setups, the three-machine flow shop problem on s sublots can be solved in O(s) time.

Doctor of Philosophy (PhD)