Abstract
This study analyses a two-stage newsvendor system with a regular and an emergency order. The emergency order can be placed at a later time based on a more accurate demand forecast. However, the unit cost for the emergency order is higher, and the quantity is limited. To maximise the expected profit, a retailer should determine both regular and emergency order quantities by considering the demand forecast updating, ordering cost and quantity constraint. Using dynamic programming, optimal ordering quantities are derived, and properties of the optimal solutions are obtained. Numerical experiments are carried out to illustrate the effect of the emergency order on the ordering decisions and expected profit. Some managerial insights are gained from the numerical results.
Acknowledgments
We are grateful to the associate editor and anonymous reviewers for their valuable comments and suggestions. The first author would like to thank Dr. Xueming Yuan for his suggestions and encouragements.