Decision timing, information inference, and information sharing: A robust supply chain game with two-way information asymmetry

This paper examines a two-echelon supply chain consisting of a retailer and a manufacturer under two-way information asymmetry, in which the two actors have private demand and cost information, respectively. In response to the uncertainty caused by information asymmetry, a robust optimization approach is used to model the risk-averse behavior of the enterprises in the game. We incorporate an information transmission mechanism by which the dominated enterprise would obtain partial private information contained in the decision of the dominant enterprise through information inference. We explore the interaction between decision timing and inference, then investigate full and partial information sharing strategies to reveal the critical roles of information flow in supply chain operations. The results illustrate the impact of demand and cost information gaps on enterprise operations, where the information gap is defined as the difference between the worst-case bound and the actual value of the demand and the cost, respectively. In addition, enterprises may not prefer sharing information when the difference between the two information gaps is large or the demand is small. Hence we propose some subsidy strategies for inducing information sharing which can achieve win-win situations. Moreover, moving first and not sharing information is a pure-strategy Nash equilibrium in most situations when not considering inference. We illustrate how the equilibrium can be changed through subsidy and inference, then provide mixed-strategy Nash equilibrium for specific situations. The results also indicate moving first is not always advantageous in the game. This depends on the follower's information gap, inference level, and information sharing strategy. Finally, in an extension, we discuss information inference costs and the impact of channel encroachment, and we validate our results based on a signaling game.