FangChen / Nanjing University of Aeronautics and Astronautics;College of Economics and Management
ZhangQing / College of Economics and Management, Nanjing University of Aeronautics and Astronautics
As the development of China's productive forces steps to the next stage, the concept of pursuing high-quality, high-innovation, high-brand-power and environment-friendly new quality productivity has come into being. Owning and improving new quality productivity has become the only way for enterprises to obtain and enhance their core competitiveness.
However, due to the high-quality and high-innovation characteristics of new quality productivity, its products always contain core technologies held by different entities, and the supply chain of new quality products is more complex than that of ordinary products. As a result, the supply chain of such products is susceptible to disruptions due to policy or competition, and companies will suffer greater losses if they fail to take effective emergency measures.
In this paper, a multi-cycle supply chain model of supplier disruptions is established. In this model, two different types of suppliers are considered, namely quality stable and price stable type. Under normal circumstances, two manufacturers purchase goods from two suppliers and sell them to different markets. In the event of supply disruptions, the two manufacturers can take additional procurement, transshipment and joint-research measures to minimize losses. This paper aims to use the Q-learning method of reinforcement learning to study the emergency strategy decision, so as to develop the optimal purchase volume, transport volume and joint-research strategy.