Professor Peyman Mohajerin Esfahani (MIE) is looking forward to teaching his undergraduate and graduate courses.
“It’s rewarding to see students develop their own way of thinking and to help them discover that mathematics is not just about formulas; it’s a powerful way of understanding and solving complex real-world problems,” he says.
A faculty member of the Department of Mechanical & Industrial Engineering (MIE), Mohajerin Esfahani approaches his course material through demonstrating how operations research and mathematical programming solve real-world problems.
Mohajerin Esfahani joined U of T in 2025, from Delft University of Technology, and is also cross-appointed with the Edward S. Rogers Sr. Department of Electrical & Computer Engineering (ECE). He brings over a decade of experience researching decision-making problems in uncertain and dynamic environments.
Kendra Hunter sat down with Mohajerin Esfahani to learn more about his research and teaching approaches.
What is your area of research expertise?
My research develops mathematical and computational tools for controlling systems and improving decision-making in dynamic and uncertain environments.
On the engineering side, this includes planning and controlling robot motion, enabling autonomous systems to operate safely, detecting anomalies in industrial processes such as high-tech printing systems before they lead to failures, and improving the efficiency of smart energy systems through technologies such as smart charging of electric vehicles.
These principles also apply to management. My research helps organizations improve supply chains and resource allocation, as well as replicate and automate human decision-making by learning from observed human behavior. I develop algorithms that combine ideas from the fields of Control Theory, Operations Research, Optimization, and Machine Learning to address these complex challenges across engineering and management.
How did you become interested in this field?
Since I was a kid, I’ve always been interested in solving puzzles. In high school, I realized that mathematics gave me the same kind of excitement. I was fascinated with the variety of ideas across different areas of math, from algebra and probability to geometry and optimization.
I really got hooked when I realized that advanced mathematics can be used to solve complex real-world problems. Even more fascinating is that the same mathematical tools often apply to problems that seem completely unrelated. For example, detecting cyberattacks or physical failures in a power grid, identifying anomalies in autonomous driving systems, and localizing leaks in water distribution networks sound like very different challenges, but mathematically they’re surprisingly similar. That realization played a big role in shaping my research interests.
What brought you to MIE?
One of my main research areas is Operations Research (OR), and the University of Toronto is one of the world’s leading institutions in this field. MIE is the primary home for OR, with another strong group at the Rotman School of Management, with a slightly different focus on Operations Management (OM).
These two groups have a close and collaborative relationship, sharing common methodological research, seminar series and student co-supervisions. One distinction between them is their application domains: MIE’s OR applies these methodologies to engineering systems, while Rotman’s OM group focuses on management and business applications. For example, at MIE, operations research and related methods are used to address challenges such as controlling and health monitoring of robotics and energy systems, as well as improving the operation and reliability of manufacturing processes. At Rotman, similar methodologies are applied to problems such as logistics, supply chain management and organizational decision-making.
Besides MIE and Rotman, I collaborate closely and co-supervise students with colleagues in ECE and the Department of Computer Science. After all, decision-making under uncertainty is a broad research area that shows up across many disciplines, so it’s common for researchers with different backgrounds to work on related problems. That’s one of the things I enjoy about being at U of T: the opportunities for interdisciplinary collaboration are everywhere.
Which courses will you be teaching this fall/winter?
In fall 2026, I’ll be teaching the undergraduate course Introduction to Machine Learning and the graduate course Markov Decision Processes. In Winter 2027, I’ll teach the new graduate course Stochastic Programming and Robust Optimization.
The main objective in Introduction to Machine Learning is essentially to “learn how to learn”. Students are introduced to the fundamental principles behind how machines learn from data, make predictions and improve their performance. They explore theoretical foundations of machine learning together with a wide range of real-world applications, from anomaly detection in engineering systems and medical diagnosis to the recent advances in large language models. The course emphasizes analytical thinking and mathematical reasoning, giving students plenty of opportunities to challenge themselves!
Markov Decision Processes is a graduate-level, research-oriented course that introduces Markov Decision Processes (MDPs), one of the fundamental mathematical frameworks for control and sequential decision-making under uncertainty. MDPs provide the theoretical foundation for many modern AI systems, including reinforcement learning, autonomous systems and intelligent decision-making algorithms. The course develops the mathematical tools needed to model, analyze and solve problems where decisions must be made over time in dynamic and uncertain environments.
Stochastic Programming and Robust Optimization is a new graduate course focusing on two powerful optimization frameworks for decision-making under uncertainty: stochastic programming and robust optimization. These approaches provide systematic tools for designing decisions that remain effective when future conditions are unknown or data are imperfect. The course covers methodologies that are widely used in applications such as logistics, energy systems, finance and engineering design, where uncertainty is an inherent part of the decision-making process.
What was your favorite course in undergraduate studies and why?
That’s a tough question! I had several favorite courses, so it’s hard to pick. But if I had to choose, I would probably say linear algebra and stochastic processes. Looking back, I can confidently say that a strong foundation in these subjects goes a long way, even at the graduate level and in research! These courses shaped the way I think about mathematical modeling and provided tools that I still use in my work today.
