Multi-agent Simulation for the transshipment problem with a non-negligible transfer lead time and a limited transportation mean capacity

We consider a supply chain consisting of n locations replenished at the beginning of each period by a supplier. These locations may coordinate in order to balance their inventory level through transshipment. Transshipment is the items transfer from location having an inventory excess to another in need. The transshipment problem consists to determine the initial inventory level where a transshipment policy is practiced. In this work, we consider the transshipment problem characterized by a non-negligible transshipment lead times and a limited transportation mean capacity. Our aim is to find a transshipment policy that reduces the inventory costs and improve the customer fill-rates. To realize this aim, we proposed a new formal transshipment model in which the period is divided into a set of sub-periods and the transshipment decision is made at the end of one of them. We also introduced a multi-agent model allowing to simulate the cooperated behavior of the inventory locations.