Comparison of supply chain replenishment strategies in a non-capacitated distribution system

This research considers inventory replenishment in a stochastic, non-capacitated, multi-echelon distribution system. Simulation is used to model Distribution Requirements Planning (DRP), reorder point and Kanban replenishment strategies. Additional experimental factors include demand seasonality, lot sizes, demand uncertainty and transit time uncertainty. Trade-off curves between inventory and delivery performance are generated. Statistical techniques, including ANOVA and regression, are then used to determine the relative dominance of the replenishment strategies by comparing the areas under the trade-off curves. Results show the methodology developed for analysis of the results works well in identifying both main and interaction effects. DRP performance is found to be the most robust, followed by that of reorder point and Kanban systems, respectively.