A simulation study of single-vendor, single and multiple-manufacturers supply chain system, with stochastic demand and two distribution policies

Abstract

This paper addresses the long-standing stochastic single-vendor, multi-manufacturer inventory control problem, using simulation optimization. It is assumed that Manufacturers producing similar goods experience very high random demand; hence, safety stock of raw materials is held in reserve in their warehouses. The vendor supplying this raw material (principal ingredients) as a policy restricts shipments to multiples of full truck load. Thus, it is necessary to take replenishment decision and coordinate delivery among these manufacturers. To solve this problem, we modeled the single vendor, single manufacturer version of the problem (AlDurgam et al., 2017) using simulation optimization techniques, which was validated numerically using parameters and results from AlDurgam et al. (2017). The simulation model was modified systematically to relax the single manufacturer assumption under two distribution policies namely, joint reorder point and vendor managed inventory. These policies were evidently modeled with stringent conditions in literature. A numerical example was provided to compare the performances of the two proposed policies, and the VMI policy was found to performed better in terms of financial savings. Lastly, we investigate the robustness of the famous continuous review (Q, R) inventory policy which is widely used in the mathematical modeling of this problem, against the common cycle assumption. The coefficient of variation is thus suggested as a judgment criterion of when to embrace simulation modeling ahead of other modeling techniques.