Design and Analysis of Data-driven Pricing and Supply Chain Strategies for Online Multi-sided Platforms

近年来，出现了许多促进买家和卖家之间点对点（P2P）交易的在线平台。这个概念与线下的概念有两个基本的不同：(i)首先，在线下空间，几乎不可能收集到关于客户如何与产品互动以及他们如何做出最终购买决策的详细信息。另一方面，消费者和零售商之间在线互动的存在使供应链公司不仅可以收集最终销售数据，还可以收集最终销售之前发生的互动链。这个机会可以帮助公司设计出定制化的战略，并适应消费者行为和属性的细节。（ii）其次，线下市场的互动是典型的单边互动。更具体地说，一边是卖方，负责运营（如采购、库存、运输）和营销（如分类、定价、促销等），另一边是客户，他们根据提供给他们的选择集做出购买决定。另一方面，在网络平台上，交易本质上是多方的。以亚马逊、阿里巴巴和Airbnb等双边平台为例。在这些情况下，平台提供商将供应商或所有者与买家或消费者进行匹配。此外，平台提供商扮演着不同的角色，从更被动的消费者和供应商之间的市场中介到更积极的角色，平台提供商代表平台参与者做出运营和营销决策。虽然之前的供应链文献中已经解决了上述一些问题，但它们都是在零碎的基础上完成的，没有充分调查市场创新对整个系统的影响。基于我目前对供应链管理（SCM）的研究，在这个为期5年的综合研究计划中，我想解决这个差距。在短期内，我开发了一个建模框架来捕捉双边市场的客户交互。利用这个框架，我打算从顾客和销售者的角度设计和分析有效和最优的销售机制，这些机制是激励相容的。从长远来看，该提案旨在为在线多边市场开发数据驱动的计算技术。在线平台是市场创新，在不久的将来将改变经济。要成功地可持续地利用这些系统的潜力，以便最有效地为社会服务，就需要采取一种综合办法，考虑到各种内部和外部行动者之间的相互依存关系。这项研究将是第一个开发这样一个综合框架，以帮助研究人员分析可持续增长的最佳机制。

In recent years, there has been an emergence of a number of online platforms facilitating Peer-to-Peer (P2P) transactions between buyers and seller. This concept differs from its offline counterpart in two fundamental ways: (i) First in the offline space, it was nearly impossible to collect detail information regarding how the customers interact with the product and how they make their final purchase decision. On the other hand, the very presence of online interactions between consumers and retailers enabled supply chain companies to collect not only the final sales data but also the chain of interactions that take place before the final sales. This opportunity helps the firms devise strategies that are customized and adapted to finer details of consumer behaviors and attributes. (ii) Second, the interactions in the offline markets are typically one-sided. More specifically, on one side is there one seller who makes operational (i.e., procurement, inventory, transportation) and marketing (i.e., assortment, pricing, promotion etc.) and on the other side are customers who are making their purchase decisions from the choice set presented to them. On the other hand, in online platforms, the transactions are inherently multi-sided. For example, consider two-sided platforms such as Amazon, AliBaba and Airbnb. In these cases, the platform providers match suppliers, or owners on one side with the buyers, or consumers on the other side. Moreover, the platform providers play different roles varying from more passive market-mediation between consumers and suppliers to more active roles in which the platform providers make operational and marketing decisions on the behalf of platform participants. While some of the above issues have been addressed before in the supply chain literature, they have been done on a piecemeal basis without fully investigating the implications of marketplace innovations for the entire system. Building upon my current research on Supply Chain Management (SCM), in this 5-year integrative research program I would like to address this gap. In the short-term, I develop a modelling framework to capture customer interactions for two-sided markets. Using this framework, I intend to design and analyze efficient and optimal selling mechanisms that are incentive-compatible from the perspectives of customers and sellers. In the long-term, this proposal intends to develop data-driven computational techniques for online multi-sided markets. Online platforms are marketplace innovations that are transforming the economy in the near future. Success in sustainably harnessing the potential of these systems so as to serve the society most effectively would require a holistic approach that takes into account interdependencies between the various internal as well as external actors. This research will be among the first to develop such an integrated framework to help researchers analyze the optimal mechanisms for a sustainable growth.