Markov decision process based multi-round negotiation in manufacturing service collaboration under dynamic pressure conditions

Due to the diverse interests and preferences of stakeholders in manufacturing service collaboration (MSC), negotiation plays a crucial role in achieving satisfactory outcomes. Although consumers and manufacturing service (MS) providers share the same negotiation context, they hold different positions and face distinct pressures from negotiation counterparts, competitors, and inherent conditions. This paper investigates the multiround negotiation in an incomplete information and dynamic pressure conditions consisting of the consumers and MS providers. First, due to changes in the number of negotiators, proposals, negotiation round, etc. in the MSC that lead to dynamic pressure conditions, a three-level pressure assessment model is constructed. Next, considering the presence of both public and private information in the MSC, a multi-attribute incomplete information game model is constructed. Finally, based on Markov decision process, an algorithm for seeking Bayesian Nash equilibrium is designed to solve this dynamic negotiation problem. Through numerical experiments, the effectiveness and superiority of the constructed model and the proposed algorithm are verified. The results show that the task completion rate is ensured, the negotiation round is remarkedly shortened, and the satisfaction of the consumers and MS providers is improved. The proposed algorithm is adaptable to most pressure conditions.