报告嘉宾

1. Guillermo GALLEGO            (香港科技大学)

报告主题：Threshold Utility Model with Applications to Retailing and Discrete Choice Models

2. Stefanus JASIN                    (密歇根大学)

报告主题：Revenue Management in the Era of Online Retail and On-Demand Business

3. Cong SHI                            (密歇根大学)

报告主题： Network Revenue Management with Online Inverse Batch Gradient Descent Method

4. Xin CHEN                            (伊利诺伊大学厄巴纳-香槟分校)

报告主题：Booking limit control in network revenue management with uncertain capacities

5. Arnoud DEN BOER              (阿姆斯特丹大学)

报告主题：Dynamic Pricing with Demand Learning and Reference Effects

Joint work-in-progress with Bora Keskin, Duke University    

6.Qi Annabelle FENG              (普渡大学)

报告主题：Operational Data Analytics:  A Newsvendor Application

7.Chung Piaw TEO                （新加坡国立大学） 

报告主题：Topic: Selected Applications of Data Driven Analytics for Pricing and Resource Allocation Problems

8.Zizhuo WANG                     (明尼苏达大学，香港中文大学 (深圳)、杉数科技)

报告主题：Applying Data Analytics for Revenue Management in China (中/英文报告)

9.Wuyang (Michael) YOU       (阿里巴巴新零售研究中心) 

报告主题：Transform of Supply Chains in New Retail Era (中文报告)

                 新零售时代的供应链变革

10.Yaoshuai ZHAO                 (中国航信研发中心新品研究所)  

报告主题：The Future of Revenue Management of China (中文报告)

                 中国收益管理的未来

11.Rui (Ray) WANG                        (阿里巴巴集团飞猪旅行) 

报告主题： Data Fusion of Revenue Management in Data Technology Era (中文报告)

                  DT时代收益管理的大小数据融合

 

           Guillermo GALLEGO           

  (Hong Kong University of Science and Technology)  

Topic: Threshold Utility Model with Applications to Retailing and Discrete Choice Models

Abstract: 

We consider a threshold utility model (TUM) where consumers buy all products whose net utility exceeds a non-negative product specific threshold. We assume that the consumer selects the thresholds to maximize the expected surplus subject to a bound on the expected number of selected products. We show that at optimality the thresholds are product-invariant. The model is arguably more appropriate  in a retail setting than a traditional discrete choice model where consumers are confined to the selection of a single alternative. After introducing and analyzing properties of the TUM, we investigate both pricing and assortment optimization under two models for the outside alternative. In the first case, the consumer embeds the outside alternative and purchases all products whose utility  exceeds the outside alternative by the common threshold. In the second case, the consumer  treats the outside alternative as a product and selects it if exceeds the threshold. We obtain optimal pricing policies for both cases. For the exponential random utility, we find that optimal prices have the product-invariant adjusted mark-up property. The assortment problem is NP-hard, but an efficient approximation is developed with tight bounds. The TUM can also be used as a two-stage discrete choice model, where a consumer first forms a random consideration set by including all the products whose net utility exceeds the threshold, and then selects the product with the highest surplus from her consideration set. We provide bounds and approximations for the expected maximum surplus and choice probabilities. We establish tight  bounds on the expected surplus of this consideration set model relative to that obtained by observing all products.

Bio of the speaker:  

Professor Guillermo Gallego is the Department Head of Industrial Engineering and Decision Analytics, and also the Crown Worldwide Professor of Engineering. 

Prior to his appointment in January 2016, Prof Gallego was the Liu Family Professor at the Department of Industrial Engineering and Operations Research at Columbia University, where he served as the Department Chairman from 2002-2008. He was named a Manufacturing and Service Operations Management Society (MSOM) Fellow in 2013, INFORMS Fellow in 2012 and has been the recipient of many awards including the Revenue Management Historical Prize (2011) and the Revenue Management Practice Prize (2012), the INFORMS Impact Prize (2016) and the Management Science Best Paper Award (2017).

Prof Gallego’s research interests are Dynamic Pricing and Revenue Optimization, Supply Chain Management, Electronic Commerce, and Inventory Theory. He has published influential papers in the leading journals of his field where he has also occupied a variety of editorial positions. His work has been supported by numerous industrial and government grants. In addition to theoretical research, Prof Gallego has developed strong collaboration with global corporations such as Disney World, Hewlett Packard, IBM, Lucent Technologies, Nomis Solutions, and Sabre Airline Solutions. He has also worked with government agencies such as the National Research Council, the National Science Foundation in United States and the Ireland Development Agency. His graduate students are associated with prestigious universities and occupy leading roles in their chosen fields. He spent his 1996-97 sabbatical at Stanford University and was a visiting scientist at the IBM Watson Research Center from 1999-2003.

Prof Gallego received both his PhD degree (1988) and MS degree (1987) in Operations Research and Industrial Engineering from Cornell University.

 

                        Stefanus JASIN                      

(University of Michigan at Ann Arbor)  

Topic: Revenue Management in the Era of Online Retail and On-Demand Business

Abstract:

How should an online retailer decide the set of products to be displayed for promotion, the set of products to be offered with a free shipping or a free one day delivery, or the set of products to be placed at the top of a display page? These are examples of so-called "product framing". It has been widely noted in the empirical literature that product framing matters. They affect customers' attention, which in turn affect their purchasing decision. This talk presents a recent joint work on the topic of randomized product framing, pricing, and order fulfillment for e-commerce retailers.  We develop a heuristic policy and numerically test our heuristic policy using both synthetic and real-world data provided by a major US retailer. The results show that the proposed heuristic is very close to optimal and also outperforms some state-of-the-art algorithms on other dimensions. Given time, I will also present another joint work on the topic of real-time dynamic pricing for balancing taxi supply and demand. This work is motivated by the new application of dynamic pricing in major taxi companies in Asia.  

Stefanus JASIN is an Associate Professor of Technology and Operations at the Ross School of Business, University of Michigan, Ann Arbor.He received his bachelor degree from UC Berkeley and masters and PhD degrees from Stanford University. Stefanus' main research interest is in developing effective and efficient approximate algorithms for tackling complex and large-scale business analytics problems using tools from statistics, machine learning, computer science, and engineering optimization. He has done works on a variety of topics including dynamic  pricing and revenue management, on-demand business analytics, online learning and optimization, and retail logistics and web analytics. 

 Cong SHI

 (Assistant Professor, Industrial and Operations Engineering, University of Michigan at Ann Arbor)
Topic: Network Revenue Management with Online Inverse Batch Gradient Descent Method

Abstract: 

We consider a general class of price-based network revenue management problems that a firm aims to maximize revenue from multiple products produced with multiple types of resources endowed with limited inventory over a finite selling season. A salient feature of our problem is that the firm does not know the underlying demand function that maps prices to demand rate, which must be learned from sales data. It is well known that for almost all classes of demand functions, the revenue rate function is not concave in the products' prices but is concave in products' market shares (or price-controlled demand rates). This creates challenges in adopting any stochastic gradient descent based methods in the price space. We propose a novel nonparametric learning algorithm termed online inverse batch gradient descent (IGD) algorithm. For the large scale systems wherein all resources' inventories and the length of the horizon are proportionally scaled by a parameter $k$, we establish a dimension-independent regret bound of $O( k^{4/5} \log k)$. This result is independent of the number of products and resources and works for a continuum action-set prices and the demand functions that are only once differentiable. Our result guarantees the efficacy of both algorithms in the high dimensional systems where the number of products or resources is large and the prices are continuous. (This is a joint work with Yiwei Chen.)

 

Bio of the speaker:

Cong SHI is an assistant professor in the Department of Industrial and Operations Engineering at the University of Michigan at Ann Arbor. His main research interests include inventory and supply chain management, revenue management, and service operations. He has won the first place in the INFORMS George Nicholson Student Paper Competition, the third place in the INFORMS Junior Faculty Interest Group (JFIG) Paper Competition, and the finalist for the MSOM Data Driven Challenge. He received his Ph.D. in Operations Research from MIT in 2012, and his B.S. in Mathematics from the National University of Singapore in 2007.

 

Xin CHEN

(University of Illinois Urbana-Champaign)  

Topic: Booking limit control in network revenue management with uncertain capacities

Abstract:

We consider the revenue management problem over a network of resources with uncertain capacities based on booking limit control. The resulting optimization model, though maximizing a non-concave objective function, can be converted to an equivalent convex program, which we solve using a stochastic gradient algorithm. We illustrate the advantage of our approach over an approach based on virtual capacity and bid-price control proposed in the literature.

Bio of the speaker:

Xin CHEN is a professor at the University of Illinois at Urbana-Champaign. He obtained his PhD from MIT in 2003, MS from Chinese Academy of Sciences in 1998 and BS from Xiangtan University in 1995. His research interest lies in optimization, data analytics, revenue management and supply chain management. He received the Informs revenue management and pricing section prize in 2009. He is the coauthor of the book “The Logic of Logistics: Theory, Algorithms, and Applications for Logistics and Supply Chain Management (Second Edition, 2005, & Third Edition, 2014)”, and serving as the department editor of logistics and supply chain management of Naval Research Logistics and an associate editor of several journals including Operations Research, Management Science, Mathematics of Operation Research and Production and Operations Management.

Arnoud DEN BOER 

(Assistant Professor, University of Amsterdam)
Topic: Dynamic Pricing with Demand Learning and Reference Effects
   Joint work-in-progress with Bora Keskin, Duke University

  

Abstract:  

We consider a seller’s dynamic pricing problem with reference effects: the phenomenon that sales is not only influenced by the current price, but also by a so-called reference price constructed in the minds of potential customers based on the seller’s price history. There is substantial empirical evidence that customers are loss averse: that means that the demand reduction when the selling price exceeds the reference price is larger than the demand increase when the selling price falls behind the reference price by the same amount. Consequently, the expected demand as a function of price has a time-varying “kink” and is not differentiable everywhere. The seller neither knows the underlying demand function nor observes the time-varying reference prices. In this setting, we show that neglecting the reference effect can be very costly. We design and analyze a policy that (i) changes the selling price very slowly to control the evolution of the reference prices, and (ii) gradually accumulates sales data to balance the trade-off between learning and earning. We prove that, under a variety of reference-price updating mechanisms, our policy is asymptotically optimal; i.e., its T-period revenue loss relative to a clairvoyant who knows the demand function and the reference-price updating mechanism grows at the smallest possible rate in T. We also extend our analysis to the case of gain-seeking customers, and show that, surprisingly,  `difficulty of the learning problem’, measured by the asymptotically optimal growth rate of the regret, is parameter-dependent. 

Bio of the speaker: 

After obtaining a mathematics degree at Utrecht University  (2006) and a post-master degree Mathematics for Industry at Eindhoven University of Technology (2008), Arnoud wrote his PhD thesis `Dynamic Pricing and Learning' (2013) at the CWI Centrum for Wiskunde and Computer Science, under the supervision of Bert Zwart and Rob van der Mei. After positions at Eindhoven University of Technology (postdoc), University of Amsterdam (postdoc), and University of Twente (postdoc / assistant professor), he joined the University of Amsterdam in 2016 as assistant professor in the mathematics department. He is also affiliated to the Amsterdam Business School. 

Arnoud's research focuses on the interface of learning and optimization, with applications in dynamic pricing and revenue management. His PhD thesis and subsequent research has been awarded the 2015 Gijs de Leve prize for best PhD Thesis in operations research defended in the Netherlands in the period 2012-2014, an NWO Veni grant in 2014 from the Dutch Science Foundation, and the INFORMS Revenue Management & Pricing Section Prize in 2016.  

Qi Annabelle FENG

 (Purdue University)

Topic: Operational Data Analytics:  A Newsvendor Application

Abstract:

With the development of computing technology and data availability, an increasing attention is paid to data-driven and data-integrated decision making in practice and research. We propose a framework of Operational Data Analytics (ODA) and demonstrate its application through the example of newsvendor model. In this talk, we focus on the situation where we have full structural knowledge but may be uncertain about the statistical characterization of the model.  The ODA framework integrates data into decision making by carefully formulating the data-integration model and the validation model. The data-integration model identifies an appropriate class of operational statistics (i.e., the decision as a statistics of the data) and the validation model finds the best performer among the operational statistics. In contrast to the traditional estimation-optimization approach, which does not optimize the actual performance measure, the ODA approach significantly improves decision quality especially when the data is limited.

Bio of the speaker: 

Qi Annabelle FENG is the John and Donna Krenicki Chair Professor in Operations Management at Krannert School of Management, Purdue University. She was previously a faculty member at McCombs School of Business, The University of Texas at Austin. She received her Ph.D. in Operations Management from UT Dallas in 2006. Her main research interest lies in studying firms’ sourcing decisions in the broad context of supply chain management. Her work focuses on individual firm’s procurement planning in uncertain environment and multiple firms’ interactions in sourcing relationships. She also works in the areas of product development and proliferation management, resource planning, economic growth models, and information system management. She is currently a Department Editor for Production and Operations Management. She received the first prize in the INFORMS Junior Faculty Paper Competition in 2009, Franz Edelman Award in 2009 and the Wickham Skinner Early-Career Research Accomplishment Award in 2012. 

 

Chung Piaw TEO

（National University of Singapore）  

Topic: Selected Applications of Data Driven Analytics for Pricing and Resource Allocation Problems

 

Abstract: 

We describe several applications of data driven analytics on pricing and resource allocation problems, building on recent advances in the field of online convex optimization. Using market shares data from price experiments, we develop techniques to learn the underlying choice model and the corresponding optimal prices. We apply this technique on data provided by a few industry partners, leading to 5%-7% improvement in profit in general. We also develop self-tuning real time resource allocation algorithm to a general class of resource allocation problems, to deal with multiple KPIs, hard constraints in resource allocation, and advanced demand information. We apply this technique on data provided by a ride hailing company, and show how the self-tuning algorithm can allocate more jobs to better drivers (increasing their earnings), without sacrificing the revenue earned by the platform and also the quality of service offered to customers.

Bio of the speaker:   

Chung Piaw TEO is Provost’s Chair Professor and Executive Director of the Institute of OR and Analytics in the National University of Singapore. Prior to his current appointments, he was a Head of Department, Acting Deputy Dean, Vice-Dean of the Research & PhD Program as well as Chair of the PhD Committee in the NUS Business School. He was a fellow in the Singapore-MIT Alliance Program, an Eschbach Scholar in Northwestern University (US), Professor in Sungkyunkwan Graduate School of Business (Korea), and a Distinguished Visiting Professor in YuanZe University (Taiwan). He is currently spearheading an effort to develop the Institute on Operations Research and Analytics, as part of the University's strategic initiatives in the Smart Nation Research Program. He study issues in service and manufacturing operations, supply chain management, discrete optimization, and machine learning . He is currently a department editor for MS (Optimization), and a former area editor for OR (Operations and Supply Chains).  He has also served on several international committees such as the Chair of the Nicholson Paper Competition (INFORMS, US), member of the LANCHESTER and IMPACT Prize Committee (INFORMS, US), Fudan Prize Committee on Outstanding Contribution to Management (China).

                Zizhuo WANG             

   (University of Minnesota and CUHK（Shenzhen), Cardinal Operations) 

Topic: Applying Data Analytics for Revenue Management in China

Abstract:

In this talk, I will share some experiences about using data analytics for pricing and revenue management in China. I will share some of the projects we involved in as well as the approaches we took. I will also discuss some interesting research topics that come out of those projects.

Bio of the speaker:  

Dr. Zizhuo WANG is an Associate Professor from the Department of Industrial and Systems Engineering (ISyE) at the University of Minnesota and the Institute of Data and Decision Analytics (iDDA) at the Chinese University of Hong Kong (Shenzhen). He received his PhD in Operations Research from Stanford University in 2012. Prior to that, he graduated from Department of Mathematical Sciences in Tsinghua University in 2007 and obtained his M.S. in Mathematical Finance in 2011 from Stanford University. His research interests mainly include optimization and stochastic modeling, especially with applications to pricing and revenue management. He has published many papers in top journal in the field of operations research and management science.  He is also the co-founder and CTO for Cardinal Operations (杉数科技), which provides data to decision solutions to various industries.

 

Wuyang  (Michael) YOU 

（ Alibaba New Retail Research Center） 

Topic：Transform of Supply Chains in New Retail Era（中文报告）

新零售时代的供应链变革

Abstract:

传统供应链假如不想被历史的潮流所抛弃的话则需要进一步的演变和进化。

1. 系统思考整体价值

无论是新零售还是传统零售，归根结底还是回归到零售本质，那就是商品、价格和服务，而链接客户的关键还是商品。如何让商品在合适的场景满足消费者，供应链人去很少真正的去从更加前端的角度去考虑。因此如何能够更加聚焦在消费者和商品，能够更高效的和前端进行协同和整合，甚至供应链不再是一个成本中心，则是未来供应链要重点关注的领域。

2. 全程数字化

零售企业打通数据孤岛，以共享服务中心构筑应用中台，以一体化数据生产线形成集团级的数据中台，利用线上线下多品牌渠道收集到的多维度数据制定“统一用户画像”，提供贯穿全生命周期的所有服务。从经营计划的设立到商品品类的设定、到商品的畅平滞划分，再到商品的选择，以及销售数量的预测、门店的自动补货、供应商智能采购、主生产计划的优化、物流计划的制定等，实现整个业态的指挥中枢的数字化，这也为未来的决策智能化奠定了坚实了基础。

3. 生态系统化

建立以客户为中心的供应链生态系统也是未来的一个趋势。

4. 灵活个性化

正如企业将商品、市场和目标消费群体进行划分一样，供应链也应该被个性化和灵活化。这种个性化是建立在多重因素之上的，包括商品特性、购买频率、成本定价、促销要求、物流属性、重要程度等。

5. 决策智能化

随着分析能力的提升，企业能从信息中获取更多的情报，建立更加智慧的供应链。应用分析能够预测哪些商品更加好卖和他们的销售数量，能建立起商品关联销售机制，能确定门店和配送中心的最优选址，能优化库存结构和建立自动补货系统，能建立缺货和爆款预警体系，能实现运输和配送成本的最优化。利用大数据、云计算、移动互联网、社交媒体、线上线下融合等实供应链现数字化和网络化。

Bio of the speaker:  

Wuyang YOU (Michael), head of Alibaba New Retail Research Center, senior expert. He joined Alibaba in 2012, and was appointed as the vice president of Ali Research Institute. His main research fields: platform model, digital transformation and upgrading of traditional enterprises, retail supply chain management and S2b platform.

游五洋 (Michael)，阿里巴巴新零售研究中心主任，资深专家。2012年加入阿里巴巴，曾任阿里研究院副院长。主要研究领域：平台模型、传统企业数字化转型升级、零售供应链管理、S2b平台。

Yaoshuai ZHAO

（China TravelSky Network Co Ltd）

Topic: The Future of Revenue Management of China（中文报告）

中国收益管理的未来

Abstract:

2000年以来，得益于中国经济的高速发展，中国旅客量从6700余万到2019年的6.1亿增长了近10倍，而从2000年南航第一家引进收益管理系统，到2014年国航第一家引入基于O&D的收益管理系统，到2019年南航系统升级，本应发挥重大作用的收益管理系统一路走来磕磕绊绊，在国内市场近乎荒废，原因很多，主题将从思想、技术、业务、管理等角度介绍中国收益管理的过去的历史、成长和不足。而近年来大数据、人工智能、云计算等新技术及航空辅助收入、动态运价、NDC等新业务形态的出现让收益管理走到一个新的风口，天生带有数据基因的收益管理系统一定能够在中国航司做强的路上发挥决定作用，新的收益管理必然是人工智能的收益管理，本主题将从观念、算法、数据和业务形态上分几部分阐述，探讨中国收益管理未来的可行性。

Bio of the speaker:  

Zhao gradutated from Xi'an Jiaotong University in 2000. Since then he has been working in China TravelSky till now as a senior engineer. He participated research and development of products such as passenger service system, e-ticketing system, revenue integrity system, revenue management system, data service etc. Since 2018, he has been responsible for the key laboratory for the application of intelligent technology of civil aviation passenger service, focusing on the application of big data and artificial intelligence of civil aviation. At present, the main research topics include the application of artificial intelligence in revenue management, dynamic freight rate and etc.

                      Rui (Ray) WANG                       

 (Fliggy, Alibaba Group 阿里巴巴集团飞猪旅行) 

Topic: Data Fusion of Revenue Management in Data Technology Era（中文报告）

DT时代收益管理的大小数据融合

Abstract:

互联网时代给酒店行业的市场竞争带来了更多的不确定性，DT时代的大数据与酒店经营的小数据在融合之时面临着信息孤岛、系统壁垒、分析时效等方面的挑战；与此同时，传统的收益管理不仅是单纯的价格影响力，除时间、空间、渠道等因素外，有效购买更需要基于用户端需求的其他要素的重构和优化组合。

Bio of the speaker:  

Rui (Ray) WANG is deputy general manager of hotel business of Fliggy, Alibaba Group, deputy secretary general of China hotel owners alliance, vice chairman of China hotel marketing alliance. He gain his doctoral degree of hotel and tourism management at Hong Kong Polytechnic University. He has worked in Hetai Hotel Group, Overseas Chinese Town Holdings Company, Singapore ATS Group, ACTION AID, ministry of science and technology of China, etc.

王睿是阿里巴巴集团飞猪酒店事业部副总经理、中国酒店业主联盟副秘书长、中国酒店营销联盟副理事长，香港理工大学酒店与旅游业管理博士生。他曾就职和泰酒店机构、华侨城集团、新加坡ATS集团、ACTION AID 国际组织、国家科技部等。