Dynamic Games: Theory, Algorithms and Applications

这项研究的目标是开发数学模型和有效的算法来解决闭环供应链、环境管理和动态定价中的实际问题。所有讨论中的问题都有两个共同的特点，使它们成为动态博弈：它们涉及少数几个相互依赖的经济主体(公司、国家)；它们本身是动态的，即主体随着时间的推移反复竞争或合作，并且存在一个积累过程，例如生产能力或污染存量。(一)供应链。供应链成员之间缺乏协调导致效率低下，即制造商和零售商的利润较低，消费者盈余较少。这项研究的目标是设计消除这些低效的机制。此外，它的目标是增加(再)制造用过的零部件，并在闭环供应链中进行回收。(二)环境管理。尽管科学家和政策制定者就处理全球环境问题的紧迫性发表了令人震惊的声明，但在控制排放方面仍然没有约束性的国际环境协议(IEA)。这项研究的目的是：(1)设计独立能源协定，包括在结构不同的国家之间形成国际能源机构；(2)制定减少排放的激励战略；以及(3)利用大规模经济-环境模型设计合作机制。(三)动态定价。在线零售业已经蓬勃发展了一段时间，过去五年的年增长率为10%。电子商务为零售商提供了随时间(以及根据消费者)轻松调整价格的可能性。本项目旨在将战略消费者纳入竞争性定价分析。这些消费者预计未来价格会下降，可以等待购买产品。如果忽视这种决策模式，零售商最终可能会损失20%至60%的收入。这项研究还涉及不透明商品的定价策略，即其属性(如酒店名称)只有在交易完成后才向买家披露的产品。本研究的概念性贡献在于发展了某些类别的动态博弈框架的存在唯一性均衡结果，这是为了考虑上述应用中的不确定性而做出的自然选择。在计算层面上，为解决应用问题而设计的算法将为在不同领域工作的研究人员所用。在应用层面上，消除供应链中的低效率对企业和对机制设计感兴趣的学者具有明显的价值。同样，确定减少污染的机制也是研究人员、政策制定者和整个社会感兴趣的问题。最后，了解动态定价有助于企业设计合适的营销和运营策略。

The objective of this research program is to develop mathematical models and efficient algorithms to solve practical problems in closed-loop supply chains, environmental management and dynamic pricing. All the problems under consideration share two features that make them dynamic games: they involve few economic agents (firms, countries) having interdependent payoffs; and they are inherently dynamic, i.e., the agents compete or cooperate repeatedly over time, and an accumulation process is present, e.g., production capacity or pollution stock. (i)Supply chains. Lack of coordination between the members of a supply chain leads to inefficiencies, i.e., lower profits for manufacturers and retailers and a lower consumer surplus. The objective of this research is to design mechanisms to eliminate these inefficiencies. Also, it aims at increasing (re)manufacturing with used parts and recycling in closed-loop supply chains. (ii)Environmental management. Despite alarming statements being made by scientists and policy makers regarding the urgency of dealing with global environmental issues, there is still no constraining international environmental agreement (IEA) on controlling emissions. This research aims to contribute to: (1) the design of IEAs, including the formation of an IEA between structurally different countries; (2) the development of incentive strategies to reduce emissions; and (3) the design of cooperative mechanisms using large-scale economic-environmental models. (iii)Dynamic pricing. The online retail industry has been booming for some time, with 10% annual growth over the last five years. E-commerce offers retailers the possibility of easily varying their prices over time (and by consumer). This project intends to bring strategic consumers into the analysis of competitive pricing. These consumers anticipate future drops in prices and can wait to purchase the product. By ignoring such a decision-making pattern, retailers may end up with a revenue loss of between 20 and 60%. This research also deals with pricing strategies for opaque goods, i.e., products whose attributes (e.g., the hotel name) are disclosed to the buyer only after the transaction has been completed. The conceptual contribution of this research is in developing existence and uniqueness equilibrium results in some classes of the dynamic games framework, which is a natural choice in order to account for uncertainty in the above-mentioned applications. At a computational level, the algorithms designed to solve the applied problems will be accessible to researchers working in varied areas. At an applied level, eliminating the inefficiencies in supply chains is of clear value to business and to scholars with an interest in mechanism design. Similarly, defining mechanisms for reducing pollution is of interest to researchers, policy makers and society in general. And finally, understanding dynamic pricing helps firms design suitable marketing and operations strategies.