TLab | Seminar 随机建模及其在城市物流、零售业与调度中的实际决策应用

- 内容摘要 -

In this presentation, we explore the application of stochastic modeling techniques to address complex, real-world decision-making challenges. Specifically, we focus on two-stage stochastic programming and stochastic sequential modeling using Markov Decision Processes (MDPs). Through practical examples in urban logistics with drones, omnichannel retailing, staff scheduling in manufacturing, and hospital operating room scheduling, we demonstrate how uncertainty can be effectively managed. Advanced solution methodologies, including Benders decomposition and reinforcement learning, are applied to tackle these large-scale stochastic models. The insights gained from these approaches offer valuable perspectives on operational efficiency, resource allocation, and cost management in uncertain environments. This talk will provide attendees with a deeper understanding of how to apply stochastic models to solve real-life problems and derive actionable insights for improved decision-making.