主讲人简介: | 王曙明,中国科学院大学经济与管理学院教授,主要从事不确定性决策与最优化、统计与优化建模、模型不确定性研究及其在选址、物流与供应链管理、交通、健康医疗管理等领域的应用。研究成果分别发表于Production and Operations Management, INFORMS Journal on Computing, Transportation Science 等权威杂志上。目前担任期刊Computers & Operations Research领域主编 (Area Editor),Decision Sciences副主编(Associate Editor)。 |
讲座简介: | We investigate several aspects of predictive prescriptions in data-driven distributionally robust optimization (DRO), including statistical modeling, performance guarantees of prescriptions, and tractability. In particular, we are trying to answer the following questions: (i) How to capture the correlation/unobservable effects in the data-driven DRO models? (ii) How to incorporate the time-series forecasts in the multi-stage DRO models? (iii) How to capture jointly the ambiguity and misspecification effects in the DRO framework. |