A demand aggregation approach for inventory control in two echelon supply chain under uncertainty

Abstract

In this paper it is discussed that the demand aggregation is an effective approach for reducing inventory levels and the number of facilities under the uncertain supply and demand conditions. Therefore in this paper, an inventory control model is developed incorporating demand aggregation approach for two staged supply chain distribution network under uncertain demand conditions. The two stage of distribution network mainly consists of distributors and retailers. This inventory control model is developed as non-linear programming model with in the different alternatives of distribution networks. The main decision variables of the system are reorder point and the ordering quantity. The prime objective function in this paper is the total cost of system which mainly consists of ordering cost, inventory carrying cost, facility cost, facility operating cost and the cost of shipment. The model is solved for total cost minimization which provides the optimum inventory policy (reorder point and ordering quantity) and the minimum cost. Through this problem best alternative of distribution network is also suggested along with optimum reorder point, ordering quantity and total cost of the system. Some other vital inventory performance parameters besides of ordering quantity and reorder point are also evaluated for the system. These performance parameters are safety stocks, expected shortages per cycle, fill rates, cycle service level, average inventory etc. These performance parameters are evaluated with total cost of the system under different uncertainty levels for a desired service level. This problem also yielded the best network options in given uncertain conditions of demand and supply. This model is formulated for single product and single period. This study mainly focused on the small part of supply chain i.e. distribution network for implementing demand aggregation approach. A case study of a sugar mill distribution network has been performed for validating the industrial applications of the proposed model.