AbstractThis paper develops, demonstrates and tests a heuristic procedure for controlling finished goods in a multi-echelon environment where there are significant transportation costs and capacity limitations on storage and transportation resources. This overall problem can be formulated as a large, mixed-integer linear programme (MILP), but solving such a model for practical-sized problems is costly and time-consuming, or even impossible, given computer systems typically used by businesses. Instead, the MILP formulation is used to both motivate and provide a benchmark for testing the performance of the heuristic procedure developed in this paper. When compared to the optimal results from the MILP, the cost performance of the heuristic procedure is found to be excellent.