Zhenghang Xu

Zhenghang Xu

He/Him

Postdoctoral Research Fellow at Rotman School of Management, University of Toronto

Zhenghang Xu

Welcome to my page!

I am Zhenghang Xu (徐正航), a Postdoctoral Research Fellow at Rotman School of Management, University of Toronto. I have the pleasure of working with Prof. Nasser Barjesteh, Prof. Sheng Liu, and Prof. Vahid Roshanaei, continuing my academic journey with the Rotman community.

I received my PhD in Operations Management and Statistics from the University of Toronto in 2026. It was my great honor and pleasure to be advised by Prof. Opher Baron, Prof. Dmitry Krass, and Prof. Philipp Afèche. Prior to my PhD, I received my bachelor’s degree in Statistical Science from The Chinese University of Hong Kong, Shenzhen (CUHKSZ) in 2021.

AI for Queueing Discovery and Control

Queueing simulators typically start with a specified model and generate customer trajectories. I study the inverse problem: learning operational dynamics from data . I develop methods that use the ordering and dependencies of queueing events to guide machine learning, evaluate counterfactual policies , and support better decisions. My research connects this discovery problem to capacity allocation and online pricing , where decisions influence congestion and the data available for future learning. I am also extending these ideas to inventory systems.

AI for Queueing Discovery and ControlOperational systems include queues and inventory. Simulation leads from systems to data; discovery leads from data to systems. Model-based control leads from systems to decisions. Data informs decisions through online learning, and decisions feed back to operational systems. Operational Systems Queue Inventory … Data Decisions Model-based control Simulation Discovery Online learning Feedback from decisions AI for Queueing Discovery and ControlQueues and inventory are operational systems. Simulation generates data; discovery learns systems from data. Online learning connects data to decisions. Model-based control connects systems directly to decisions, and decisions feed back to systems. Operational Systems Queue Inventory … Data Decisions Simulation Discovery Online learning Model-based control Feedback to systems

Queueing Discovery

I use the structure of queueing event data to construct sufficient candidate inputs for machine learning models and evaluate counterfactual policies.

Featured paper: Queueing Causal Models.

Inventory extension: SCIM (ongoing research).

Capacity Control

I use fluid models and optimal control to allocate scarce capacity across patient classes in persistently overloaded healthcare systems.

Featured project: MRI Capacity Allocation.

Online Learning

I design pricing policies that learn customer values while accounting for feedback between prices, admissions, congestion, and future observations.

Featured paper: Learning to Price in Queues.

Seminar and Contact

I co-organize the Rotman Young Scholar Seminar series. Please feel free to contact me about the seminar or potential research connections.

Contact: zhenghang.xu@rotman.utoronto.ca