Optimal schedule for extended basic period approach of economic lot scheduling problem

Abstract

Most algorithms for the economic lot scheduling problem (elsp) following the extended basic period approach consist of two decision levels. On the upper level, the length of the production cycle and the number of lots (frequency) within the cycle for all products are determined. On the lower level, lots are scheduled to level workloads of all periods and ensure a timely start of production. This paper presents a new mixed-integer programming (mip) model for the scheduling subproblem under the power-of-two policy. This is the first mip model that exactly determines and minimizes additional inventory holding costs due some lots’ premature start of production. It may be solved by a free general-purpose solver within a fraction of a second. Experiments with several problem instances described in the literature confirmed that using the new model within a heuristic algorithm ensures a significant cost reduction for the entire elsp. Additionally, all optimal schedules for the Bomberger case are presented.