管理科学系学术讲座

时   间：2023年12月12日（周二） 13:30-15:00

地   点：管理学院思源楼524室

主   题： Integrating Prediction and Optimization: A Practical End-to-End Inventory Management Model

主讲人：戚萌 康奈尔大学助理教授

主持人：虞嘉怡 复旦大学管理学院副教授

摘    要：

Big data provides new opportunities to tackle one of the main difficulties in decision-making: uncertain behaviors. Instead of the classical two-step predict-then-optimize (PTO) procedure, we provide data-driven solutions that integrate these two steps. I will first discuss a multi-period inventory replenishment problem with uncertain demand and vendor lead time (VLT), with accessibility to a large quantity of historical data. Different from the traditional two-step predict-then-optimize (PTO) solution framework, we propose a one-step end-to-end (E2E) framework that uses deep-learning models to output the suggested replenishment quantity directly from input features without any intermediate step. The E2E model is trained to capture the behavior of the optimal dynamic programming solution under historical observations, without any prior assumptions on the distributions of the demand and the VLT. This algorithm is currently implemented in production at JD.com to replenish thousands of products. The concept of integrating prediction and optimization is useful for a wide range of circumstances beyond supply chain management. Therefore, if time permits, I will move to a more general setting of the contextual stochastic optimization problem. We propose an integrated conditional estimation-optimization (ICEO) framework that estimates the underlying conditional distribution using data while considering the structure of the downstream optimization problem. We demonstrate desired theoretical performance guarantees and discuss the computational efficiency.

简    介：

Meng Qi is an Assistant Professor of Operations, Technology and Information Management at Cornell University SC Johnson College of Business. Her research focuses on data-driven solutions for new models of revenue and supply chain management in modern e-commerce. Specifically, her research provides both methodologies and practical solutions combining tools and concepts from optimization, machine learning, and statistics. Meng received her Ph.D. in Industrial Engineering and Operations Research from UC Berkeley and her B.S. in Mathematics and Physics from Tsinghua University.

管理科学系

2023-12-6