A scheduling problem in a group technology manufacturing cell that exhibits flow shop characteristics is presented. The problem involves scheduling n jobs on m machines where the jobs are assigned to an undetermined number of cells; all cells have a common fixed cycle time that should not be exceeded by the completion time of the last job assigned to each cell, and there is no buffer between machines. The objective is to minimize the sum of avoidable idle times for the machines and waiting timesfor the jobs. A heuristic algorithm has been developed that also applies to agile manufacturing systems. Because no other algorithm has been developed to solve this problem, a random scheduling (RS) algorithm and generic simulated annealing (SA) algorithm have been developed to solve the problem for comparison purposes. Experiments conducted indicate that when the proposed algorithm is compared to the RS and SA algorithms, it reduces the total avoidable idle times for the machines and waiting times for the jobs by up to 50%, it reduces the number of cells needed to achieve a given cycle time by 10%, and it reduces the CPU time needed by up to 90%.
展开▼
