The mixed integer-linear models that have been used in most analyses of warehouse location problems fail to capture the potential operating efficiencies associated with large scale facilities. This paper presents an algorithm for finding a minimal cost warehouse system design wherein individual warehouses have limited capacities and exhibit economies of scale. The iterative procedure defines and solves a series of conventional transportation problems in order to converge on the optimal system design. The algorithm is well suited to solving large problems of the type commonly encountered in practice.