An Efficient Algorithm for Mixed Model Just-in-Time Production System with Chain Constraints

Abstract

There has been growing interest in scheduling problems where the jobs are penalized both for being early and for being tardy. A mixed model manufacturing facility running under a just-in-time production system is controlled by setting the production schedule for the highest level in the facility, which is usually a mixed model final assembly line. The schedule is set to achieve the goals of the organization, which under JIT are to keep a constant rate of part usage and to maintain a smooth production load. We consider the former goal in this dissertation.

This dissertation includes different literature as well as the recent trends in JIT environment. Our concern in this dissertation is to find out the possible optimal sequences for controlling JIT production system for mixed-model production systems with Chain Constraints and min-sum deviation objective. For this, we consider non-overlapping chains, and by considering each chain as a pseudo job and their length as demands, we can have a pseudo schedule from EDD, which is later replaced by the real job, can lead a combined optimal chain sequence.

Therefore, in this case, an optimal sequence can be obtained in efficient time complexity. Our results extend the previous results on non-overlapping chain sequences with absolute-deviation objective function.