Mahdi Mostajabdaveh

Mahdi Mostajabdaveh

Senior Staff Scientist

AI & Optimization

I am an Operations Research and AI scientist with over a decade of experience designing algorithms—metaheuristics, exact optimization, and machine learning methods—to solve large-scale, real-world business problems.

Currently, I am a Senior Staff Scientist at the Huawei Vancouver Research Center and the AI Lead for the Huawei Optimization Solver (OptVerse). I lead research on integrating machine learning and large language models into a commercial-grade solver, from algorithm design and benchmarking to usability and productization. Our work powers applications such as large-scale vehicle routing, logistics planning, and network design.

Previously, I was a Postdoctoral Research Fellow in the Department of Mathematics and Industrial Engineering at Polytechnique Montréal and CIRRELT, supervised by Prof. Michel Gendreau and Prof. Teodor Crainic. I received my Ph.D. in Industrial Engineering and Operations Management from Koç University.

My work spans both fundamental research and applied innovation, leading to patents, peer-reviewed publications (NeurIPS, ACL, AAAI, EJOR, C&OR), and deployed optimization solutions.

Research & Professional Interests

AI for Optimization & Solvers

Learning-augmented heuristics and branch-and-bound, LLM-based optimization modeling and agentic workflows, GPU-accelerated metaheuristics, and intelligent cut and search strategies to make solvers faster, more robust, and easier to use.

Applied Optimization in Industry

Using OR and ML to design scalable solutions for last-mile and middle-mile logistics, production scheduling, and power distribution, with a focus on models and algorithms that can be deployed in real systems.

Selected Projects

Tool-Augmented LLMs for Operations Management

SmartAPS is an agentic conversational system that transforms how operations planners interact with Advanced Planning Systems. Using tool-augmented LLMs with retrieval-based API selection, it enables natural language what-if and why-not scenario analyses, reducing consultant dependency from days to hours.

Generative AI For Optimization Modeling

More details coming soon.

HybridMind: AI-Human Collaboration for Algorithmic Ideation

More details coming soon.

View All Projects

Selected Publications

AI for Operations Research

Evaluating LLM Reasoning in the Operations Research Domain with ORQA

M Mostajabdaveh, TTL Yu, SCB Dash, R Ramamonjison, JS Byusa, et al.

Proceedings of the AAAI Conference on Artificial Intelligence (2025)

Optimization Modeling and Verification from Problem Specifications Using a Multi-Agent Multi-Stage LLM Framework

M Mostajabdaveh, TT Yu, R Ramamonjison, G Carenini, Z Zhou, Y Zhang

INFOR: Information Systems and Operational Research (2024)

Multi-Agent System Optimization

MASPRM: Multi-Agent System Process Reward Model

M Yazdani, M Mostajabdaveh, Z Zhou, Y Xiong

arXiv preprint (2025)

Real-World Applications

Inequity-Averse Shelter Location for Disaster Preparedness

M Mostajabdaveh, WJ Gutjahr, F Sibel Salman

IISE Transactions (2019)

A Branch-and-Price Algorithm for Fast and Equitable Last-Mile Relief Aid Distribution

M Mostajabdaveh, FS Salman, WJ Gutjahr

European Journal of Operational Research (2025)

Methodological Advancements in OR

Bayan Algorithm: Detecting Communities in Networks Through Exact and Approximate Optimization of Modularity

S Aref, M Mostajabdaveh, H Chheda

Physical Review E (2024)

Analyzing Modularity Maximization in Approximation, Heuristic, and Graph Neural Network Algorithms for Community Detection

S Aref, M Mostajabdaveh

Journal of Computational Science (2024)

View All Publications

Recent Blog Posts

Disaster and Equity

Examining the role of equity in disaster management and resource allocation.

View All Posts

Contact

Feel free to reach out for collaborations, research opportunities, or just to connect.