A mobile ammunition distribution system design on the battlefield

Ankara : The Department of Industrial Engineering and the Institute of Engineering and Science of Bilkent University, 2010.Thesis (Ph.D.) -- Bilkent University, 2010.Includes bibliographical references leaves 150-158.Ammunition has been the most prominent factor in determining the outcome of combat. In this dissertation we study a military logistics problem in which ammunition requirements of the combat units, which are located on the battle- field, are to be satisfied in the right amount when and where they are needed. Our main objective is to provide a decision support tool that can help plan ammunition distribution on the battlefield. We demonstrate through an extensive literature review that the existing models are not capable of handling the specifics of our problem. Hence, we propose a mathematical programming model considering arc-based product-flow with O(n 4 ) decision variables and constraints. The model is a three-layer commodity-flow location routing formulation that distributes multiple products, respects hard time windows, allows demand points to be supplied by more than one vehicle or depot, and locates facilities at two different layers. We then develop a new mathematical programming model with only O(n 3 ) decision variables and constraints by considering node-based product-flow. We derive several valid inequalities to speed up the solution time of our models, illustrate the performance of the models in several realistically sized scenarios, and report encouraging results. Based on these mathematical models we propose two three-phase heuristic methods: a routing-first location-second and a location- first routing-second heuristic. The computational results show that complex real world problems can effectively be solved in reasonable times with the proposed heuristics. Finally, we introduce a dynamic model that designs the distribution system in consecutive time periods for the entire combat duration, and show how the static model can be utilized in dynamic environments.Toyoğlu, HünkarPh.D