The design of production-distribution networks: A mathematical programming approach

This text proposes a mathematical programming approach to design international production-distribution networks for make-to-stock products with convergent manufacturing processes. Various formulations of the elements of production-distribution network design models are discussed. The emphasis is put on modeling issues encountered in practice which have a significant impact on the quality of the logistics network designed. The elements discussed include the choice of an objective function, the definition of the planning horizon, the manufacturing process and product structures, the logistics network structure, demand and service requirements, facility layouts and capacity options, product flows and inventory modeling, as well as financial flows modeling. Major contributions from the literature are reviewed and a number of new formulation elements are introduced. A typical model is presented, and the use of successive mixed-integer programming to solve it with commercial solvers is discussed. A more general version of the model presented and the solution method described were implemented in a commercial supply chain design tool which is now available on the market.