Evaluation of centralized distribution system in clothing industry with risk-pooling approach (a case study)

As supply chains and its distribution centres expand, risk increases and can have very different impacts on the optimal inventory management policies to meet consumer demand. The basic idea of risk pooling in supply chains is that if the customer demands from all sources are pooled by a centralized system, then not only will expected profits increase, but also inventory levels and costs will decrease. Regarding to customer demand variability of textile clothing products, efficient inventory management in clothing supply chain leads to total inventory cost reduction, and then retaining competitive edge. In this paper, the effect of customer demand uncertainty in multiple warehouses system under the two centralized and decentralized systems is taken into consideration in the presence of deterministic supply. In order to investigate the effect of risk pooling and its result under two scenarios in (s,S) inventory control namely, 1) stochastic demand 2) stochastic demand and lead-time, a two-echelon clothing supply chain is taken into account under the aforementioned scenarios. We consider customer demand variability in a single manufacturer, multiple distribution centers and multiple retailers (demand points). We also assume that inventory is stored at only a single echelon, distribution centers, in order to make decisions about centralization versus decentralization of supply chain under consideration. In the latter part of this paper, aggregate warehouse is then located by logistical costs minimization through a real case study. In doing so, different transportation modes, their related costs and distances from the centralized distribution center are involved by comparing potential location sites for the obtained pooled inventory center. Numerical results show that if customer demand distribution for products is normal, average inventory in centralized distribution system is less than that of decentralized distribution systems. Finally, sensitivity analyses are numerically conducted to investigate critical inventory parameters on service level of the local company.