| An analysis of control policies for a two-stage supply chain with subcontractors at each stage is presented when decisions at each stage concerning safety stocks, backorders, and subcontracting are made jointly or in a decentralized manner. The inventory/admission control policies considered are base stock, echelon base stock and partial backordering, and the objective is to maximize the mean profit rate of the system. The optimal control parameters are found by exhaustive search using Markov chains. From numerical examples it appears that the policies which manage jointly sales and production levels in each stage provide much higher overall (system) profits than decentralized policies, although the latter are individually more profitable for the second stage. Partial back ordering is slightly more beneficial than no backordering or infinite backordering, offering extra profit. Finally we investigate also the sensitivities of the optimal profit rates and control parameters to changes in the parameters of the system. |