This paper proposes a mathematical model for optimal delivery plan and inventory strategy in
distribution system and suggests a solution approach. Distribution system consists of single distributor and
multiple retailers. For example, distributor can be a logistic center and retailer can be Walmart or Kroger.
Retailers face uncertain and changing demand. The problem we have to solve is to make decisions about
which retailers, when, and how many inventory to deliver. If small inventory is allocated to the retailer
which has large demand, lost sales or back orders may happen. On the other hand, if a number of
inventory is allocated to the retailer which has only a few demand, the inventory holding cost will
increase. This paper deals with the problem that has long lead time and a limited amount of inventory in
distributor. Thus we have to deliver the inventories to multiple retailers with optimal plan. In addition,
since the product has short life cycle and is perishable, we consider the transshipment. In this paper, we
develop a two stage stochastic programming model and suggest a solution method which uses scenarios for uncertain demand and Benders decomposition.