Abstract Usually each manufacturing stage in a supply chain makes its own decision regarding quantity and timing of parts it purchases from its suppliers, thereby controlling its inventory position and overall supply chain dynamics. Such decisions, although good for each individual stage, can adversely affect the overall performance of the supply chain. This can be viewed as distributed control of inventory, in which each controller is making autonomous decisions based on local objectives. Because these controllers do not have information about inventory position or order quantities at other stages, the safety stock tends to be higher, leading to higher inventories and cost. This also causes demand amplification and the bullwhip effect. This paper presents a distributed feedback control algorithm, called the Adaptive Logistics Controller (ALC), which simultaneously decides the order quantities for each stage of the supply chain to minimize the total WIP in the entire supply chain for a given demand. In this approach, simulation is used to provide the feedback to the ALC controller, leading to an iterative numerical computational approach. Computational experiments compared with the traditional centralized ( Q , r ) policy model show that the order quantities calculated by the distributed ALC are much superior in terms of total overall WIP and hence result in lesser total costs for the entire supply chain.