Saunak Kumar Panda

I am a Postdoctoral Scholar at the Human-Centered AI Institute in the C.T. Bauer College of Business at the University of Houston. I received my PhD in Industrial Engineering, with a specialization in Operations Research, from the University of Houston in 2025, where I worked with Dr. Yisha Xiang.

My research focuses on developing reinforcement learning and statistical methods for sequential decision-making under uncertainty in engineering and operational systems. My work spans risk-sensitive learning, statistical inference, optimization, and AI-assisted decision support, with applications in manufacturing process control, resource allocation, personalized healthcare recommendations, and procurement auctions.

Research Interests

  • Reinforcement learning
  • Stochastic modeling
  • Risk-sensitive sequential decision-making
  • Statistical inference and uncertainty quantification

Application Areas

  • Manufacturing and industrial process control
  • Healthcare and treatment recommendation
  • Resource allocation
  • Procurement auctions

News & recognition

Oct 2025

Finalist, INFORMS RAS Data Challenge

Our team was among six international finalists for railway wheel failure prediction.

Sep 2025

Joined the Human-Centered AI Institute

Started as a Postdoctoral Scholar at the University of Houston.

Aug 2025

Completed my PhD in Industrial Engineering

University of Houston, with a specialization in Operations Research.

Oct 2024

Presented at INFORMS Annual Meeting 2024

Online Statistical Inference of time-varying sample-averaged Q-learning

Aug 2024

Presented at Reinforcement Learning Conference 2024

Online Statistical Inference of sample-averaged Q-learning

Oct 2023

Presented at INFORMS Annual Meeting 2023

Online Statistical Inference for dynamically-changing batch Q-learning

Oct 2022

Presented at INFORMS Annual Meeting 2022

Solving Dynamic Resource Matching in Manufacturing using Reinforcement Learning

May 2022

Presented at IISE Annual Conference & Expo 2022

Dynamic matching of demand-supply types with manufacturing resources