Real-time order acceptance in transportation under uncertainty

Abstract

Without using any order acceptance criteria, retail companies distributing products with private transportation fleets are not able to maximize their profits because they are not adequately utilizing their capacity. The objective of this paper was to create and validate a model to determine if historical demand data can be used by retail firms operating private fleets to make effective real-time order acceptance/rejection decisions with the purpose of eliminating unprofitable orders in a short-haul transportation setting. A Java tool was generated to instantaneously decide whether or not to accept an order depending on the order location and time of receipt. The model was tested against optimal decisions using total demand knowledge and several alternative real-time decision-making strategies. The model was found to significantly outperform the alternative real-time decision making strategies and provided profits approximately eight percent lower than the optimal decisions. We conclude that using historical demand probabilities is useful in informing the decisions of retail firms seeking to utilize private fleets efficiently and increase profitability through cost reduction.