Integrated scheduling of production and logistics operations of a multi-plant manifacturer serving a single customer area

Abstract

Increasing market competition forces manufacturers to continuously reduce their leadtimes by minimizing the total time spent in both production and distribution. Some studies in the relevant literature indicate that scheduling production and logistics operations in a coordinated manner leads to improved results. This thesis studies the problem of scheduling production and distribution operations of a manufacturer serving a single customer area from multiple identical production plants dispersed at different geographical locations. The products are transported to the customer area by a single capacitated truck. The setting is inspired by the operations of a leading soft drink manufacturer, and the objective is set in line with their needs as the minimization of the total completion time of the jobs. The completion time of a job is defined as the time it reaches at the customer area. We consider both this general problem and four special cases motivated by common practical applications. We prove that both the main problem and three of its special cases are NP-hard at least in the ordinary sense. We develop mixed integer programming (MIP) models for all these problems and propose a pseudo-polynomial dynamic programming mechanism for the remaining special case. Since the MIP models are able to provide optimal solutions only for small instances in a reasonable amount of time, heuristics are also proposed to solve larger instances. Fast lower bounds are developed to facilitate the performance assessment of these heuristics in medium and large instances. Evidence from extensive computational experimentation suggests that the proposed heuristics are both efficient and effective.