Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


PhD_Thesis(2).pdf (5.54 MB)

ETD Abstract Container

Abstract Header

Operation of Multi-modal Transportation Networks: Optimization and Reinforcement Learning

Deng, Yuntian
http://rave.ohiolink.edu/etdc/view?acc_num=osu1657501083332872

Abstract Details

2022, Doctor of Philosophy, Ohio State University, Electrical and Computer Engineering.
Myriad options and modes are now available for passengers to commute between different places. At the moment, all these services are owned and operated by distinct competitors, and every firm is trying to attract as many customers as possible. Such uncoordinated competition and myopic operation result in a huge cost to society in terms of serious traffic congestion, high energy consumption, and low utilization of public resources. In this dissertation, we propose new models, algorithms, and theoretical analyses to overcome the observed challenges through optimization and reinforcement learning. Our results provide insights into the pricing of multi-modal networks and mode-specific decisions. In particular, we consider the following problems: Pricing of the multi-modal network when demand function is known or unknown: We consider the situation where multiple transportation service providers cooperate to offer an integrated multi-modal platform. The platform sets incentives (price discounts or excess charges on passengers) along every edge of the transportation network. When the demand function is known, the optimal incentives that maximize the profit of the platform are obtained through a two-time-scale stochastic approximation algorithm. When the demand function is unknown and time-varying, we leverage kernelized bandit to learn the nonparametric demand function and maximize the system profit. We consider both service constrained and unconstrained cases and use restarting or weighted method to tackle non-stationarity. Once the profit is determined, we use the asymmetric Nash bargaining solution to design a fair profit-sharing scheme among the service providers and show that each provider's profit increases after cooperation on such a platform. Real-time and strategic decisions on ride-hailing system: For real-time decisions, we focus on designing a rebalancing algorithm for a large-scale ride-hailing system with asymmetric demand. We pose it within a semi-Markov decision problem (SMDP) framework and minimize a convex combination of the passenger's waiting time and the total empty vehicle miles traveled. We use a deep reinforcement learning algorithm to determine the approximately optimal solution to the SMDP. For strategic decisions, we focus on fleet sizing and charger allocation in the electric vehicle sharing system. We develop a closed queueing network model to analyze the performance given a certain number of chargers in each neighborhood. Depending on the demand distribution, we devise algorithms to compute the optimal fleet size and number of chargers required to maximize profit while maintaining a certain quality of service. We further show that two slow chargers may outperform one fast charger when the variance of charging time becomes relatively large in comparison to the mean charging time.
Ness B. Shroff (Committee Chair)
Abhishek Gupta (Committee Co-Chair)
Qin Ma (Committee Member)
Xingyu Zhou (Committee Member)
Eylem Ekici (Committee Member)
253 p.
Electrical Engineering; Operations Research

Recommended Citations

Citations

  • Deng, Y. (2022). Operation of Multi-modal Transportation Networks: Optimization and Reinforcement Learning [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1657501083332872

    APA Style (7th edition)

  • Deng, Yuntian. Operation of Multi-modal Transportation Networks: Optimization and Reinforcement Learning. 2022. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1657501083332872.

    MLA Style (8th edition)

  • Deng, Yuntian. "Operation of Multi-modal Transportation Networks: Optimization and Reinforcement Learning." Doctoral dissertation, Ohio State University, 2022. http://rave.ohiolink.edu/etdc/view?acc_num=osu1657501083332872

    Chicago Manual of Style (17th edition)

osu1657501083332872
224
© 2022, some rights reserved.Creative Commons License Operation of Multi-modal Transportation Networks: Optimization and Reinforcement Learning by Yuntian Deng is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by The Ohio State University and OhioLINK.