Warehouse location under multiple transportation options.

摘要

The deregulation of the U.S. freight transportation industry has increased significantly the spectrum of transportation options available to shippers, thereby increasing the importance of representing multiple transportation options in the design and analysis of a logistics system. The objective of this study is to develop a methodology for analyzing warehouse location under multiple transportation options. The proposed methodology recognizes and represents the interdependence between facility location, transportation, and inventory decisions.; The problem of warehouse location under multiple transportation options, termed the Combined Warehouse Location-Transportation Problem (CWLTP), is defined as that of determining the number and locations of warehouses, and the "optimal" transportation options between plants and warehouses, such as to minimize total distribution cost. The proposed CWLTP model differs from existing warehouse location models in three important aspects. First, it considers the selection of transportation options as output rather than input. Second, it includes an explicit representation of the inventory implications of warehouse location and transportation decisions. Thirdly, it explicitly represents the required level of customer service.; The CWLTP is a complex mixed integer problem which cannot be solved directly using existing optimization techniques. We develop a heuristic algorithm for solving the CWLTP, which is based on decomposing the CWLTP into two subproblems. Each of the two subproblems is shown to be a component of the CWLTP. With appropriate simplifications, the first subproblem is reduced to a linear mixed integer problem, and is solved exactly. The second subproblem is solved exactly using an implicit enumeration scheme.; We show that the CWLTP solution algorithm provides valid solutions to a sample of test problems of different sizes, and produces reasonable responses to changes in company policy and in the environment in which the company operates. It is also shown that a simultaneous optimization of warehouse location and transportation decisions can lead to a significantly lower total distribution cost, than a component-by-component approach in which warehouse location and transportation decisions are determined independently.