DESIGN OF LARGE-SCALE LOGISTICS SYSTEMS FOR UNCERTAIN ENVIRONMENTS

In this research, the author proposes a continuum approximation methodology developed for deterministic problems that can be used in analyzing uncertain logistics systems. Using this methodology, cost approximations are obtained which improve with scale and near-optimal configuration of systems is achieved. The proposed methodology considers two basic logistics system design problems. The first is the standard vehicle routing problem (VRP) in which customer locations and demands are uncertain when planning, and the size of the load that can be transported by the vehicle constrains the tours. The second problem is a time-constrained VRP in which customers require uncertain service times and vehicles are required to return to the depot by a time deadline. Both traditional and newly proposed system designs are analyzed and compared for both problems.