I 摘要 随着信息技术的发展和互联网的普及，零售商通过线上线下多渠道分销以实 现销售体系的打通。线上和线下不再是泾渭分明的两个领域，两者的结合，才能 推进电子商务的快速发展。2013年，天猫和银泰集团合作，线上和线下同步销售， 探索O2O（Online to Offline）模式。2015年，美团和大众点评的合并，将O2O推 向了风口浪尖。京东商城与实体店永辉超市进行深度合作来实现O2O。天猫“双 十一”狂买后，接踵而来的是大量退货。据此，在O2O供应链系统中，产品如何 捆绑销售以及有效减少渠道间产品退货等渠道策略一直是企业界和学术界亟待解 决的现实难题。在“线上到线下”的O2O模式中，面对消费者需求迁移的渠道转 换行为，零售商如何选择销售渠道来提升供应链绩效是值得研究的新课题。苏宁 易购和海尔集团通过网上购物平台销售新产品，同时通过线下实体门店销售再制 造品。因此，探讨新产品/再制造品销量以及博弈方的经济效益、环境可持续性和 消费者剩余具有很强的现实意义。 优衣库推出的线上支付线下取货的模式，满足了消费者的购物体验，带动了 线下的客流。GameStop公司通过线下渠道销售加工处理后的再制造产品，并在线 提供最新游戏软件。基于此，本文围绕电子商务企业的现实背景，综合运用主从 博弈理论、供应链管理理论和最优化理论等，拓展了O2O供应链渠道选择领域的 研究，为解决企业渠道销售问题提供具体的、可操作的技术与方法。区别于已有 研究，创新性的考察了捆绑销售、消费者需求、产品退货、协调机制和再制造品 分销成本对O2O供应链系统的影响，并进行数值分析验证模型的实用性，为企业 管理者提供理论依据和决策参考。 本文主要创新性研究工作有四个方面构成，总结如下： （1）研究由一个供应商和一个平台商组成的二级供应链系统，通过线上和线 下渠道独立或捆绑销售两种产品。构建了两个不同的渠道结构模型，采用逆向归 纳法推导出博弈方的最优批发价格和零售价格以及各自的最大利润。研究发现： 当市场规模不大时，捆绑销售是占优策略；当市场规模超过一定范围时，独立销 售是占优策略。无论谁作为主导方，整条O2O供应链利润相等。无论是供应商还 是平台商作为Stackelberg主导者，自身价格敏感系数与博弈方的最大利润呈单调 递减关系；独立销售策略下，交叉价格敏感性系数与博弈方的最大利润呈单调递 增关系。 （2）线上线下需求迁移成为常态，影响着博弈方的决策。研究由一个平台商， 摘要 II 一个实体店，一种产品构成的O2O供应链系统。平台商采取线上直销渠道，线下 渠道采取自营门店的集中决策或加盟门店的分散决策。构建了需求无迁移和迁移 两种情形下的O2O供应链渠道结构模型，设计出分散决策下的协调机制以优化 O2O供应链。研究表明，无论自营门店还是加盟门店，系统总利润增加，线上线 下售价的高低取决于需求迁移比例的范围；随着线上迁移比例的增加，线下售价、 系统总利润提高，线上售价、分散决策时平台商利润降低；当固定转移支付费用 在一定范围内时，两部定价契约可以有效的协调O2O供应链。 （3）消费者喜欢“触摸和感觉”来体验产品是否匹配，零售商在计算退货预 期成本时，需考虑提供这种服务水平产生的成本。基于不同的分销渠道，讨论了 零售商的退货预期成本对博弈方经济效益的影响，构建了双头垄断和完全垄断情 景下的O2O供应链渠道结构模型，以获取批发价格和销售价格，以及博弈方的最 大利润。研究表明：当退货预期成本在一定范围内时，双头垄断情景下的供应商 和O2O供应链的利润（模型D）高于完全垄断情景下的利润（模型O）。然而，无 论退货预期成本如何，平台商的利润始终高于两个零售商的利润之和。对于消费 者剩余，当退货预期成本满足一定条件时，在完全垄断情景下（模型O）的消费 者剩余低于双头垄断情景（模型D）。 （4）线上平台零售商和线下实体零售商之间的竞争普遍存在于商业活动中， 涉及两种情形下的O2O供应链渠道结构模型：① 模型P：制造商提供一种产品（包 括新产品和再制造品）给两个零售商，其中新产品通过线上平台（零售商R1）销 售，再制造品通过线下实体店（零售商R2）销售；② 模型M：制造商将此产品 批发给一个线上和线下渠道同时经营的零售商R销售。研究发现：从经济视角看， 与模型P相比，模型M提高了零售商的经济绩效，但削减了制造商和产业的利润； 从环境的持续性看，模型P比模型M更环保；从社会视角看，模型P的消费者剩 余总是高于模型M，再制造品的分销成本越高，越不利于消费者。 本文创新性的对电子商务环境下的O2O供应链渠道选择问题展开系统性研究， 提供了全新的研究视角，深化了O2O供应链渠道选择领域的研究，具有理论上的 前沿性和探索性，为企业管理者提高运营效益提供理论基础和应用指导。 关键词：O2O供应链，捆绑销售，需求迁移，产品退货，再制造 ABSTRACT III ABSTRACT With the development of information technology and the popularity of the Internet, online and offline multi-channel distribution is used by retailers to achieve the integration of sales system. Online and offline are no longer two distinct fields. Only by combining them can the rapid development of e-commerce be promoted. In 2013, Tmall cooperated with Yintai group to conduct synchronous online and offline sales and explore the O2O (Online to Offline) model. In 2015, the merger of Meituan and Public Comment pushed O2O to the forefront. Jingdong mall carries out in-depth cooperation with yonghui supermarket, a physical store, to achieve O2O. After Tmall’s “Double Eleven” was bought wildly, followed by a large number of returns. Accordingly, in an O2O supply chain system, how to bundle products and how to effectively reduce product returns between channels have been practical problems to be solved urgently in the business and academic circles. In the “online-to-offline” O2O model, how to choose the sales channel to improve the supply chain performance through two-line operation is a new topic worth studying in the face of channel switching behaviors of demand shift. Suning and Haier Group sell new products through online shopping platforms, while selling remanufactured products through offline physical stores. So, it is of great practical significance to explore the quantity of new / remanufactured products as well as the economic benefits, environmental sustainability and consumer surplus. Uniqlo launched the online payment and offline pick up mode, which satisfies the consumer’s shopping experience and drives the offline passenger flow. GameStop sells processed remanufactured products through offline channel and provides the latest game software online. Based on this, around the practical background of e-commerce enterprises, the research in the field of O2O supply chain channel selection is expanded by the comprehensive application of the principal-subordinate game theory, supply chain management theory and optimization theory. It is of great theoretical and practical significance to provide specific and operable technologies and methods for solving business channel sales problems. This dissertation is different from existing research, innovatively investigated the impact of bundling, consumer demand, product return, coordination mechanism and remanufactured product distribution cost on O2O supply chain system, and carried out numerical analysis to verify the practicability of the model, ABSTRACT IV and the research results provide theoretical basis and reference for business managers. The main innovative research work consists of four aspects, which are summarized as follows: (1) This dissertation studies a two-echelon supply chain system consisting of a supplier and a platform vendor, which sells two kinds of products independently or in a bundle through online and offline. Two different channel structure models are constructed, and the optimal wholesale price and retail price and the maximum profit of the players are derived by the reverse induction method. It is found that when the market size is not large, bundling is the dominant strategy; When the market size exceeds a certain range, independent sales is the dominant strategy. Regardless of who the Stackelberg leader is, the whole O2O supply chain receive equal profit. For a market leader, independent sales or bundling decisions should be made according to market size. The self-price sensitivity coefficient and their profit decrease monotonously for both independent sales and bundling; For independent sales, as the cross-price sensitivity coefficient increases, the profit monotonically increases. (2) The mutual shift of offline and online demand becomes the norm, affecting the decision-making of the players. Research on the O2O supply chain system consisting of a platform vendor, a physical store, and a product. The platform vendor adopts direct selling online, centralized decision-making of a self-operated store or the decentralized decision-making of a franchised store offline. The O2O