SOLVING THE ECONOMIC LOT SCHEDULING PROBLEM WITH IDENTICAL FACILITIES IN PARALLEL USING GENETIC ALGORITHMS

In this study, we solve the Economic Lot Scheduling Problem (ELSP) for a production system with identical facilities in parallel. The ELSP with identical facilities in parallel is concerned with the lot sizing, scheduling, and production assignment decision of n items so as to minimize the total costs per unit time. Since the ELSP with identical facilities in parallel is NP-hard, we propose two three-phase solution approaches based on the Genetic Algorithm (GA). In the first phase, we employ either Carreno's [5] heuristic or a GA to determine the assignment of n items to the identical facilities in parallel. Then, another GA utilizes the advantage of its multi-directional search ability to search for the candidate solutions (i.e., the replenishment cycles of the products) in the second phase. The third phase uses an efficient heuristic to test the feasibility of the candidate solutions and tries to generate a feasible production schedule for each facility. Based on our random experiments, our GA-based approaches are able to efficiently solve the ELSP with identical facilities in parallel within a reasonable run time, and their solution quality dominates Carreno's [5] heuristic.