A Risk-Reward Competitive Analysis for the Newsboy Problem with Range Information

Abstract

Recently, the single-period, single-item newsboy problem with limited distributional information (e.g., range, mean, mode, variance, symmetry) has been widely studied. However, the existing newsboy models with partial information are only fit to risk-neutral inventory managers. This paper considers the newsboy problem with range information. Based on the competitive ratio analysis, which guarantees a certain performance level under all possible input sequences, we construct a framework to manage risk and reward of newsboy problems under different forecasts (i.e. certain forecasts; probability forecasts; probability distributions). Comparing the existing studies, this approach helps the newsboy flexibly choose the optimal reward strategies, according to his own risk tolerance levels and different forecasts.