Abstract

Trucking plays a critical role in the economy. Recent startups are disrupting the industry with mobile technology, creating app-enabled freight platforms. Known as Uber for trucks, these platforms connect truck drivers to more loads and provide more flexibility compared to traditional brokers. We explore a dynamic pricing problem on these platforms where truck drivers shop through available loads. We formulate the problem as a Markov decision process (MDP) that captures randomness on both sides of the market and incorporates drivers’ choice behavior. As MDPs are challenging to solve in general, we study two approximation frameworks: 1) an approximate dynamic program and 2) a fluid model. These approximations are tractable and their solutions provide managerial insights. Numerical experiments show that heuristics based on these approximations achieve near-optimal revenue performance.

Time

2019年12月23日（星期一）10:00-11:30

Speaker

Yufeng Cao is a Ph.D. candidate in Operations Research at Georgia Institute of Technology. He works on revenue management and marketplace analytics with applications in the airline, logistics, and retail industries. His research strives to narrow the gap between customer behavior modeling and empirical evidence and develop efficient methods to improve business operations. Yufeng received his bachelor's and master's degrees in Electrical Engineering from Tsinghua University and had consulting experience with companies both in China and in the US.

Venue

信息管理与工程学院308室

上海财经大学（第三教学楼西侧）

上海市杨浦区武东路100号