Conveying Demand Information in Serial Supply Chains with Capacity Limits

For serial multiechelon systems subject to production capacity limits at every stage, we consider a class of modified echelon base stock ( MEBS ) policies. To evaluate information requirements of such systems, we consider two separate inventory management mechanisms operated in a decentralized manner. For ordering decisions, these mechanisms utilize local knowledge only and are distinguished by the timing of the orders being conveyed upstream from installation to installation. We demonstrate that these mechanisms can duplicate the shipment quantities in the modified echelon base-stock policy that uses full information. Thus, although full demand information will not be conveyed up the channel due to the demand censoring effects of capacity, we demonstrate that sufficient information about the market demand is conveyed via the orders. This suggests that local information is sufficient to make ordering decisions that replicate the policy’s orders, a significant finding for implementing supply chain inventory policies in practice, where dynamic state information may not be readily accessible. We extend this local information result to serial channels with completely general capacity configurations acting under the corresponding echelon policies. We demonstrate the strong relationship between these two mechanisms that relate to serial capacitated channels of differing lengths. We augment our main results with two important extensions. (1) Because our focus is on MEBS policies, which are not necessarily optimal for longer supply chains, we evaluate their performance. We numerically show that they perform very well in general. We also provide upper and lower bounds, which further justify their strong performance. Given that these policies are close to optimal, that they are easy to interpret, and that they can operate with only local information, they are appealing in practical applications. (2) We compare the usage of local information to operate a capacitated system versus incentives to sustain such a system. We show that, similar to the noncapacitated case, it is possible to design an alternative incentive-compatible performance mechanism such that local managers will follow the centralized solution, albeit with more demanding information requirements.