Fulfillment flexibility strategy for dual-channel retail networks

Abstract

Flexibility design has been widely adopted in practice as a competitive strategy to respond effectively to uncertainties. In this article, we analyze the fulfillment flexibility for dual-channel retail networks, in which the firms should fulfill both online demands from retailing platforms and in-store offline demands. In particular, by setting the order of fulfillment, we find that a dual-channel retail network can be equivalently transformed to an online retail network with stochastic inventory and demand. By implementing copositive programming, we obtain an asymptotic robust lower bound for the ratio of expected sales to fully flexible expected sales under a K-chain design. This bound only depends on the partial moment information and support set of demands, rather than the complete demand distribution information. Interestingly, we derive the optimality of a K-chain in symmetric balanced networks and the performance of the K-chain under different distributions is robust. In addition, numerical experiments are conducted to further deliver some insights for practitioners. The uncertainties of in-store demand or inventory will reduce the expected sales while both fulfillment flexibility and safety inventory can be used to enhance the performance of a retail system. Finally, we find that the correlation coefficient between in-store demand and online demand will affect the decision-making significantly.