Analysis of Supply Contracts with Commitments and Flexibility

Y. Bassok and R. Anupindi

*** Abstract ****

In this paper we address an important class of supply contracts called the Rolling Horizon Flexibility (RHF) contracts. Under such a contract a buyer has to forecast and commit requirements for components, for each period, at the beginning of the horizon. Usually, a supplier provides some flexibility to adjust the current order and future commitments in a limited way in a rolling horizon manner. We present a general model for the study of RHF contracts. We propose several heuristics and derive a lower bound. We demonstrate, numerically, the effectiveness of the heuristics for both stationary and non--stationary demands. Some of the heuristics are easy to compute and hence amenable to practical implementation.

We propose two measures for the order process which allow us to (a) evaluate the effectiveness of RHF contracts in restricting the variability in the orders, and (b) measure the accuracy of advance information vis a vis the actual orders. We demonstrate, numerically, that the order process variability decreases significantly as flexibilities decrease without significantly increasing the costs to the buyer. Our numerical studies provide several other managerial insights for the buyer; e.g., we are able to provide insights into how much flexibility is sufficient, the value of additional flexibility, the effect of flexibility on customer satisfaction (as measured by fill rate), etc.