The finite horizon economic lot sizing problem in job shops : the multiple cycle approach

This paper addresses the multi-product, finite horizon, static demand, sequencing, lot sizing and scheduling problem in a job shop environment where the planning horizon length is finite and fixed by management. The objective pursued is to minimize the sum of setup costs, and work-in-process and finished products inventory holding costs while demand is fulfilled without backlogging. We propose a new and efficient cyclic scheduling solution framework, called the multiple cycle (MC) method, based on the assumption that the cycle time of each product is an integer multiple of a basic period. This method relies on a decomposition approach which decomposes the problem into an assignment sub-problem, a sequencing sub-problem and a lot sizing and scheduling sub-problem. To evaluate its performance, the MC method was compared to the common cycle method and numerical results show that it performs better, as expected. However, the magnitude of improvement varies between 4% and 8% depending on the structure of the problems.