华中科技大学管理学院

国内外高端学术论坛

(暨第四届电子商务供应链优化论坛)

会议简介

随着数字技术的快速发展，数字化已成为中国经济发展的新引擎。基于数据驱动的决策、优化作为产业发展的关键，发展潜力巨大，受到社会的广泛关注。为了加强管理科学领域学者之间的学术交流，促进同行之间的学术研究水平的提升，华中科技大学管理学院于2021年5月28日–6月29日采用线上线下相结合的方式举办以数字化运营策略为主题的第四届电子商务供应链优化论坛。本论坛旨在为电子商务供应链优化领域的海内外专家学者搭建一个学术交流平台，分享产学研经验以及电子商务供应链优化领域的新挑战和前沿研究成果。

    我们热情邀请您出席本次论坛！

会议议程

报告专家简介

吴俊杰，北京航空航天大学经济管理学院教授、副院长、网安学院双聘教授，国家自然科学基金委杰出青年基金获得者。获全国百篇优秀博士学位论文、中国电子学会技术发明一等奖、中国商业联合会科技进步一等奖。担任北航“数据智能研究中心”主任，国务院学位委第八届学科评议组（管理科学与工程学科）成员，国家自然科学基金委重大研究计划（大数据驱动的管理与决策）指导专家组成员，中国管理科学与工程学会“人工智能技术与管理应用”分会主任委员，《系统工程理论与实践》、《中国管理科学》编委/责任编委。长期从事管理科学、信息科学、社会科学的交叉创新研究，主要研究兴趣为数据挖掘、复杂网络分析、社会/城市/金融计算，在短文本建模、异质大数据融合建模、可解释的预测建模等领域取得了突出的成绩。近年主持并完成国家自然科学基金重点、科技部863大数据、工信部国家信息安全等项目30余项，在Springer发表英文专著1本，在SCI/SSCI检索国际顶级期刊如ISR、TKDE、DMKD发表论文超过30篇，在计算机领域顶级会议KDD、IJCAI、AAAI等发表论文超过20篇，1篇论文入选ESI十年高被引论文，获国家发明专利授权3项。

报告题目：用户细粒度偏好感知的深度推荐系统

摘要：近年来，推荐系统已经逐步成为了新兴电子商务平台的核心技术之一。但当前的推荐系统仍然存在无法处理冷启动产品、难以适应数据的稀疏性、对用户偏好理解不深等问题。本次报告将探讨如何基于表征学习的框架来建模现实世界中广泛存在的隐式和显式产品关系，从而为解决上述推荐系统的长期难题提供有益的思路。

李文立，大连理工大学经济管理学院教授、博士生导师，第八届国务院管理科学与工程学科评议组成员，教育部新世纪人才荣誉获得者，教育部电子商务教指委委员，辽宁省电子商务教指委主任委员，辽宁省、西藏自治区电子商务专家，中航工业科技委顾问，系统工程学报、中国管理科学编委。主持国家自然科学基金重点项目1项、面上项目5项、国家重点研发计划子课题1项。在国内外重要期刊ISR、EJOR等以及国际会议发表论文130余篇。其主要研究方向是电子商务和信息管理。

报告题目：社交资本对企业信用的影响机理研究

摘要：参照个人行为纳入个人信用评价的方法，我们考虑将企业的社交行为作为企业信用评价的补充，重点在于探讨企业社交行为如何影响企业的信用。本研究以企业社交互动形成的社会资本为切入点，探析其在企业信用评价中的作用。从理论上建立并实证检验了一个基于社会资本理论的企业社交活动对信用的影响模型。使用来自14,544家企业的数据，发现企业的社交活动(提问题、发帖和评论)对企业信用有显著影响。在企业提问、企业发帖的数量与企业信用的关系中，度中心性和中介中心性起到中介作用。该研究扩展了与企业信用相关的非财务因素作用，为全面评价企业信用提供了一个补充。

刘心报，合肥工业大学管理学院教授,国家杰出青年科学基金获得者，享受国务院政府特殊津贴专家，国家自然科学基金委创新研究群体负责人，高等学校学科创新引智基地（111计划）负责人。现任合肥工业大学校长助理、研究生院常务副院长，兼任国务院学位委员会第八届管理科学与工程学科评议组成员、中国系统工程学会智能制造系统工程专业委员会主任委员。主要从事决策科学与技术、智能制造工程管理等方面的研究工作。先后主持国家863项目、国家自然科学基金创新研究群体项目、国家自然科学基金重点项目、国家重点研发计划项目课题等国家级科研项目以及企事业单位委托的重要项目20余项，发表学术论文150余篇，在Springer出版社出版英文专著《Optimization and Management in Manufacturing Engineering》1部。近五年来，在《Production and Operations Management》《Omega》《European Journal of Operational Research》等国际期刊上发表SCI期刊论文51篇。获国家科技进步二等奖2项、教育部自然科学一等奖1项、中国机械工业科技进步一等奖1项、安徽省科技进步一等奖1项，获国家级教学成果奖二等奖1项。

报告题目：智能互联环境下制造资源协同优化理论与方法

摘要：制造资源协同优化对于提升制造企业竞争力具有重要意义，智能互联环境下制造资源协同优化理论与方法发生了变革。本报告介绍了团队研发的“马鞍山钢铁公司离散制造企业生产优化与执行系统”、“长发铝业有限公司生产调度系统”、“航天某院多项目资源协同优化系统”，以及在工程实践中提炼的工件恶化情形下生产与维修协同调度问题、生产运输协同调度问题、多技能多模式资源受限的多项目调度问题的研究过程；阐述理论与实践双向互动并相互促进的科研理念。

朱未名，博士毕业于马里兰大学（University of Maryland），现担任IESE商学院运营管理助理教授，并于麻省理工学院担任客座教授。其主要研究方向包括共享经济以及供应链金融。主要研究课题包括网约车司机行为分析，平台的产品展示优化以及定价决策，平台电商在供应链金融中的风险承担角色以及基于数据的零售商销量预测等。曾为摩拜，滴滴出行，阿里巴巴，京东，百分点科技，NCC Media以及JDA Consulting提供数据分析以及咨询服务。其研究成果发表于国际顶级期刊Management Science, Manufacturing & Service Operations Management, Journal of International Economics，并荣获多个顶级国际期刊（M&SOM，POMS）最佳论文奖。

Title: Estimating and Exploiting the Impact of Photo Layout: A Structural Approach

Abstract: Host-generated property images as a visual channel reveal substantial information about properties. Selecting proper images to display can lead to higher demand and increased rental revenue. In this paper, we define, estimate and optimize the impacts of Airbnb photos on customers' renting decisions. We apply Resnet50, a convolutional neural network model, to build two separate, supervised learning models to evaluate the image quality and room types posted by Airbnb hosts. Then, we characterize the overall impacts of photo layout by the room type featured in the photo, photo quality and the order of display on the listings' webpages. To address two estimation challenges in the Airbnb setting, namely censored demand and changing consideration sets, we propose a novel pairwise comparison model that utilizes customers' booking sequence data to consistently estimate the impact of photo layout on customers' renting decisions. Our estimation results suggest that the cover image has a significantly larger impact than non-cover photos and that a high quality bedroom cover image leads to the largest increase in demand. Furthermore, we build a non-linear integer programming optimization problem and develop an algorithm to determine the optimal photo layout. Our counterfactual analysis suggests that a listing's unilateral adoption of optimal photo layout leads to 11.0% more bookings on average. Moreover, depending on the neighborhood and market size, when listings simultaneously switch to the optimal photo layout, they get booked for two to five additional days in a year on average, which boosts the revenue by $500 to $1100.

孙建坤，本科就读于清华大学工业工程系，博士毕业于美国西北大学(Northwestern University)的Kellogg管理学院，现任伦敦帝国理工学院商学院运营管理助理教授。其研究兴趣是数字平台运营，特别是数字化和人工智能如何重塑一个组织的运营和影响消费者行为，运用数据分析和理论建模技术研究平台运营中的实际问题。其研究成果发表在Management Science和Manufacturing & Service Operations Management等国际权威期刊上。

Title: Predicting Human Discretion to Adjust Algorithmic Prescription: A Large-Scale Field Experiment in Warehouse Operations

Abstract: Conventional optimization algorithms that prescribe order packing instructions (which items to pack in which sequence in which box) focus on box volume utilization yet tend to overlook human behavioral deviations. We observe that packing workers at the warehouses of Alibaba Group deviate from algorithmic prescriptions for 5.8% of packages, and these deviations increase packing time and reduce operational efficiency. We posit two mechanisms and demonstrate that they result in two types of deviations: (1) information deviations stem from workers having more information and in turn better solutions than the algorithm; and (2) complexity deviations result from workers' aversion, inability or discretion to precisely implement algorithmic prescriptions. We propose a new "human-centric bin packing algorithm" that anticipates and incorporates human deviations to reduce deviations and improve performance. It predicts when workers are more likely to switch to larger boxes using machine learning techniques and then pro-actively adjusts the algorithmic prescriptions of those ``targeted packages.'' We conducted a large-scale randomized field experiment with the Alibaba Group. Orders were randomly assigned to either the new algorithm (treatment group) or Alibaba's original algorithm (control group). Our field experiment results show that our new algorithm lowers the rate of switching to larger boxes from 29.5% to 23.8% for targeted packages and reduces the average packing time of targeted packages by 4.5%. This idea of incorporating human deviations to improve optimization algorithms could also be generalized to other processes in logistics and operations.

张文昌,本科毕业于清华大学，获得了数学和物理学士双学位，在加州大学伯克利分校(University of California, Berkeley)获得统计学硕士学位，在马里兰大学(University of Maryland)的Robert H.Smith商学院获得运营管理博士学位，现任印第安纳大学凯利商学院商业分析学的助理教授。其研究方向主要是将实证方法和理论相结合，研究在线市场的设计和运营。其研究重点是管理市场厚度、信息披露和在线市场扩张。其研究成果发表在Management Science等国际权威期刊，也获得了多个奖项，包括IBM服务科学学生论文竞赛(2017)的冠军。

Title: Information Provision in Two-Sided Platforms: Optimizing for Supply

Abstract: While information design has gained significant attention in the recent literature as a tool for shaping consumers' purchase behavior, little is known about its use and implications in two-sided marketplaces, where both supply and demand consist of self-interested strategic agents. In this paper, we develop a dynamic game-theoretic model of a two-sided platform that allows for heterogeneity and endogenous behavior on both sides of the market. We focus on illustrating the potential benefits of optimal information provision in terms of managing supply-side decisions, including supplier entry/exit and pricing. Our analysis identifies three distinct mechanisms through which information design may increase platform revenues. First, when the outside options available to consumers and service providers are relatively unattractive, information design can be used to mimic the so-called "damaged goods" effect, allowing the platform to fine-tune its composition of providers and achieve a more revenue-efficient matching between supply and demand. Second, when consumers and/or providers have access to relatively attractive outside options, information design can help the platform increase its transaction volume significantly; interestingly, we find that in order to ramp up its throughput, the platform may need to understate the quality of its best providers. Third, when the platform uses commission subsidies to resolve the "cold-start" problem and incentivize the entry of new providers, information design can help achieve the same goal while extracting higher commission revenues; thus, we highlight the role of information design as a substitute for commission subsidies. Overall, our numerical experiments suggest that, by influencing supplier decisions, optimal information provision can lead to a substantial increase in platform revenues.

    刘晟，本科就读于清华大学工业工程系，并于2019年获得加州大学伯克利分校运筹学博士学位，博士期间参与了Amazon和Lyft的研究工作，现在就职于多伦多大学管理学院，担任运营管理和统计学助理教授。他目前的主要研究兴趣包括智慧城市和数据驱动的决策模型。其研究成果发表在Management Science和Operations Research等国际权威期刊上。

报告题目：物流系统中的数据驱动模型

摘要：本次报告旨在分享两个物流系统中数据驱动模型的应用研究。第一个研究通过轨迹数据学习外卖骑手的配送行为，并依此建立优化模型来提高配送系统的分单效率。第二个研究提出了基于机器学习的配送时间分布预测模型，以及优化目标配送时间的决策框架。两个研究都在真实数据上得到了充分验证。

崔若濛，本科毕业于清华大学工业工程系，于2014年在美国的西北大学(Northwestern University)的Kellogg管理学院获得运营管理博士学位，现在美国排名前20的埃默里大学(Emory University)任职教授。担任管理领域顶级期刊《Manufacturing and Service Operations Management》和《Production and Operations Management》的高级主编。同时也任职于澳大利亚研究委员会发现项目（Australian Research Council Discovery Projects）的咨询委员会，担任香港研究资助局（Research Grants Council of Hong Kong）的研究项目评审专家。曾担任10多个国际会议的会议主席和程序委员会主席，20次担任顶级国际论文大赛裁判官，担任国际学术组织INFORMS Junior Faculty Interest Group的财务主管。崔教授的主要研究方向有运营战略、经济学、数字化转型的理论及应用，给阿里巴巴，百度，亚马逊等多家公司提供过战略战术分析咨询。在国际知名的管理领域顶级期刊发表论文15余篇，其中管理科学期刊《Management Science》6篇，制造和服务运营管理期刊《Manufacturing and Service Operations Management》4篇，生产和运营管理期刊《Production and Operations Management》1篇，哈佛商业评论《Harvard Business Review》2篇，商业案例2篇。Google Scholar的论文总引用数达到559，h-index为11。研究荣获多个奖项，包括2020年阿里巴巴创新研究奖，2020年最佳社会意义研究奖（2020 M&SOM Society Award for Responsible Research），2019年 INFORMS青年教授最佳论文竞赛奖（2019 INFORMS Junior Faculty Interest Group Paper Competition），2019年INFORMS服务科学最佳论文奖（2019 INFORMS Service Science Section Best Paper Competition），2019年 M＆SOM基于实践的最佳论文竞赛奖（2019 M&SOM Practice-Based Paper Competition），2017年INFORMS行为科学管理最佳论文奖（2017 INFORMS Behavioral Operations Management Section Best Working Paper），和2014年POMS供应链管理学生论文竞赛奖（2014 POMS Supply Chain Management Student Paper Competition）。

Title:  Promising Delivery Speed in Online Retail

Abstract:  Online retailers have to provide customers with an estimate of how fast an order can be delivered before they purchase it. Retailers can strategically adjust this delivery speed promise online without changing offline infrastructure, and it may fundamentally impact business outcomes. It can influence consumers' purchasing decisions and post-purchase experiences, often in the opposite direction. On one hand, an aggressive (i.e., faster) delivery estimate could ensure that more customers meet their deadlines and thus may increase their purchases ex ante. On the other hand, an aggressive estimate tends to overpromise customers, risking a longer than expected wait time, which can lower customer satisfaction and increase product returns ex post. In this research, we study the causal effect of retailers' delivery speed promise on customer behaviors and business performance. Collaborating with Collage.com, an online retailer that sells customized photo products across the US, we exogenously varied the disclosed delivery speed estimates online while keeping the physical delivery speed unchanged. Using a difference-in-differences identification and a dataset with 212,340 transactions in 7,090 cities, we find that a one-day faster promise increases sales by 0.73%, profits by 2.0%, and value per order by 3.5%; a one-day slower promise reduces sales by 0.51%, profits by 2.7%, and value per order by 3.1%. However, the aggressive disclosure increases product returns and deteriorates customer retention. Our findings provide managerial insights that retailers could leverage to customize their delivery promises.

    赵经隆，本科毕业于清华工业工程系，博士就读于麻省理工学院，现在波士顿大学(Boston University)任职教授。其研究方向主要是研究优化和计量经济学之间的接口，利用离散优化技术来解决大数据时代的数字实验设计问题。其研究成果发表在Management Science等国际权威期刊上。

Title: Design and Analysis of Switchback Experiments

Abstract: In switchback experiments, a firm sequentially exposes an experimental unit to a random treatment, measures its response, and repeats the procedure for several periods to determine which treatment leads to the best outcome. Although practitioners have widely adopted this experimental design technique, the development of its theoretical properties and the derivation of optimal designs have been elusive. In this paper, we establish the necessary results for practitioners to apply this powerful class of experiments with minimal assumptions. Our main result is the derivation of the optimal design of switchback experiments under a range of different assumptions on the order of the carryover effect --- that is, the length of time a treatment persists in impacting the outcome. We cast the optimal experimental design problem as a minimax discrete optimization problem, identify the worst-case adversarial strategy, establish structural results, and solve the reduced problem via a continuous relaxation. For switchback experiments conducted under the optimal design, we provide two approaches for performing inference. The first provides exact randomization based p-values, and the second uses a new finite population central limit theorem to conduct conservative hypothesis tests and build confidence intervals. We further provide theoretical results when the order of the carryover effect is misspecified. For firms that possess the capability to run multiple switchback experiments, we also provide a data-driven procedure to identify the likely order of the carryover effect. We conduct extensive simulations to study the empirical properties of our results, and conclude with some practical suggestions.