报告一题目：
Dynamic Assortment with Online Learning under Threshold Multinomial Logit Model
报 告 人：南京大学工程管理学院 沈厚才 教授
报告时间：2025年9月12日（周五） 9:00-10:30
报告地点：文管学馆B306
报告摘要：
Consumers often find themselves overwhelmed by extensive assortments offered by retailers and therefore may exhibit bounded rationality in their purchase decisions. However, existing dynamic assortment optimization literature typically assume customers make rational choices from all available options. This motivates us to employ a simple but effective two-stage consider-then-choose(CTC) model, namely the Threshold Multinomial Logit (TMNL) model to investigate the dynamic assortment optimization problem under bounded rationality. The TMNL model assumes that a consumer first forms an endogenous consideration set by the threshold effect, then makes a choice from the consideration set, which can capture more flexible substitution patterns than the classical MNL choice model. We propose a novel two-phase online assortment optimization policy that learns consumers' CTC choice behavior. In the first phase, our policy learns the endogenous consideration set formation through online identification. In the second phase, our policy transforms the consideration set formation into a graph structure and adaptively adjusts parameter learning methods based on maximum connected components and shortest path analysis to achieve nearly optimal regret bounds. We establish that no online learning algorithm can work without our first-phase structural learning approach, and classical UCB and elimination-based algorithms fail to achieve competitive regret without our adaptive estimation mechanism. Overall, our proposed method offers a viable pathway for designing online assortment algorithms tailored to CTC behavior. Extensive numerical experiments comprehensively validate the efficacy of our proposed algorithms.
报告人简介：
沈厚才，南京大学工程管理学院教授、博士生导师，同时兼任南京大学数学系运筹学与控制论专业教授、博士生导师。他于1986年在南京大学数学系获得理学学士学位，于1989年、1995年在东南大学获得自动控制理论及应用专业硕士与博士学位。2001年回到南京大学之前，他分别在南京师范大学政教系、东南大学经济与管理学院工作。他的主要研究工作是运用数据、模型等分析方法去研究包括物流运营系统的管理与控制。他的研究成果分别发表在Production and Operations Management、IEEE Transactions on Automatic Control、Naval Research Logistics、European Journal of Operational Research、Decision Sciences、管理科学学报、自动化学报等国内外学术期刊上。目前他是澳门科技大学商学院访问教授、江苏省运筹学会副理事长、江苏省管理科学与工程学科联盟副理事长、中国管理现代化研究会运作管理专业委会委员、中国管理科学与工程学会供应链与运营管理分会委员会委员、INFORMS会员、Production and Operations Management Society会员，Modern Supply Chain Research and Applications期刊Associate Editor，《工业工程》期刊领域主编，曾任苏宁易购集团股份有限公司独立董事(2013-2020)。
报告二题目：
Service Oriented Considerate Routing: Data, Predictions and Robust Decisions
报 告 人：南京大学工程管理学院 陈彩华 教授
报告时间：2025年9月12日（周五）10:30-12:00
报告地点：文管学馆B306
报告摘要：
In this research, we focus on improving service oriented routing by addressing the nuanced challenge of punctuality through the consideration of couriers’ ability to ensure on-time deliveries. We utilize a comprehensive real-world dataset from a cold chain logistics firm for analysis and highlight critical elements, including couriers’ fixed effects and workload, as key covariates to improve prediction performance. Then we integrate couriers’ workload and location familiarity into our Service Oriented Routing (SOR) model to enhance predictions of delivery times. We introduce the Courier Assigned Location Mismatch (CALM) metric as a less intrusive approach to incorporating couriers’ location familiarity into their delivery efficiency. We propose the novel Service Oriented Considerate Routing (SOCR) model. By minimizing the CALM metric, couriers are assigned routes within familiar territories to the extent possible within the total routing distance constraint. Additionally, we develop the connection of the SOCR model with a robust satisficing approach. To solve the SOCR model, we apply Benders decomposition for an exact solution and Tabu Search for a heuristic approach, demonstrating their effectiveness and superior out-of-sample performance.
报告人简介：
陈彩华，国家优秀青年基金获得者、国家自然科学基金重大项目课题负责人、美国斯坦福大学访问学者，现任南京大学教授、博士生导师、工程管理学院副院长、民建江苏省委大数据与人工智能委员会主任。南京大学理学博士，新加坡国立大学联合培养博士。从事数据驱动的（不确定）决策、人工智能驱动的优化决策、大规模优化算法设计与应用等研究，代表作发表/接受在Mathematical Programming、UTD24系列期刊、SIAM系列杂志、IEEE系列杂志及人工智能顶级会议NeruIPS、CVPR等，获江苏省科学技术进步奖、华人数学家联盟最佳论文奖(2017、2018)，中国运筹学会青年科技奖(2018)，南京大学青年五四奖章(2019)，南京大学青年五四奖章集体（2024）。
欢迎有兴趣的师生积极参加！