Stochastic Control Through a Modern Lens: Applications in Supply Chain Analytics and Logistical Systems

The thesis investigates classical multi-period stochastic control problems through a modern lens, including stochastic inventory control, dynamic pricing and vehicle routing. We revisit vehicle routing, dynamic pricing and inventory control i) in a data-driven fashion; ii) with flexible architectures. First, we consider the vehicle routing problem with stochastic demands (VRPSD). By combining ideas from vehicle routing and manufacturing process flexibility, a new approach to VRPSD is proposed, that uses overlapped routing with customer sharing in route determination, whose performance is close to the theoretical lower-bound, and significantly improves on a routing strategy without overlapped routes. We then study the following question: how much data is needed in order to obtain a (nearly) optimal policy for joint pricing and inventory control? We propose the first sample-efficient solution for this problem. Next, the same question for inventory control is considered, and a novel sample-based algorithm is proposed for the backlog setting and a matching (up to a logarithmic factor) lower bound is also given. Finally, we conclude and point out several future research directions.

Thesis Committee:

Prof. David Simchi-Levi (Thesis Advisor), CEE/IDSS/ORC, MIT

Prof. John Tsitsiklis, EECS/IDSS/ORC, MIT

Prof. Cathy Wu (Chair), CEE/IDSS, MIT