A JOB-SPLITTING HEURISTIC FOR LOT-SIZE SCHEDULING IN MULTISTAGE, MULTIPRODUCT PRODUCTION PROCESSES

This paper focuses primarily on cyclic scheduling in multi-stage, mult i-product production. Cyclic schedules are generalizations of common c ycle schedules (CCSP) in which the cycle time for each product is requ ired to be an integer multiple (not necessarily one as in the CCSP) of the base cycle time. Solution procedures which have been reported in literature for this problem ignore in-process inventory and are limite d to single-stage and/or single-product production processes. After re viewing the literature, this paper presents an efficient heuristic pro cedure which generates near-optimal cyclic schedules for the multi-pro duct, multi-stage flow shop production environment under deterministic and stationary conditions. This procedure accounts for all inventory costs including in-process inventory and idle time. The proposed solut ion is then compared to the optimal CCSP solution developed by El-Najd awi and Kleindorfer and, as expected, performed significantly better t han the traditional CCSP approach. Compared to other heuristics studie d in the literature, this procedure appears to be more useful as a com putable and cost-efficient approach for practical size applications.