Predictive analytics for infrastructure systems

Current/past projects:
• Physics-constrained graph neural networks for transportation network modeling
• Resilience assessment of mobility networks using graph neural networks
• Explainable-AI for long-term traffic prediction
• Post-disaster online prediction of traffic conditions
Learn more

Optimization and control using AI/ML

Current/past projects:
• Structural topology optimization under uncertainty using variational autoencoders
• Geometry-aware neural operator for design optimization
• Computational control of power systems with uncertain distributed energy sources
• Deep learning for infrastructure asset management

Learn more

Physics-informed machine learning

Current/past projects:
• Physics-informed neural operators for design optimizations
• Physics-informed variational autoencoders for solving stochastic differential equations
• Pioneering enhanced training algorithms for physics-informed neural networks

Learn more

Intelligent transportation systems

Current/past projects:
• Multifidelity traffic estimation using data from multiple sources
• Efficient data collection algorithm for connective vehicles with precision guarantees

Learn more

Available Research Assistant Position

Our group has several openings in Fall 2025 for PhD students interested in machine learning and its applications on the modeling and optimization of engineering systems. Higher consideration is given to those who have background in deep learning, numerical methods, and probability and statistics and have excellent programming skills. Interested students should send an email to meidani (at) illinois (dot) edu and attach their CV including their graduate coursework.

Recent News

  • [Sep 2023] Our collaborative project with Fermilab on using graph neural networks for neutrino experiments are funded.
  • [Sep 2023] Our submission to ICMLA 2023 Conference on Graph Pyramid Autoformer for Long-Term Traffic Forecasting has been accepted. This was a collaboration with Argonne National Lab and Lawrence Berkeley National Lab. Congrats to Vincent for leading the effort!
  • [Aug 2023] Vincent will present two papers on traffic prediction at 2023 INFORMS Annual Meeting.
  • [Oct 2022] Our submission to NeurIPS Workshop on Machine Learning and the Physical Sciences has been accepted. Congrats to Rini Gladstone and Mohammad Nabian!
  • [Oct 2022] Our group is invited to present at Nvidia's introductory webinar on Physics-ML Link to the free webinar. . Rini will present our collaborative work with Nvidia on how to enforce exact boundary conditions for PINNs.
  • [May 2022] Rini is selected as a Research Scientist Intern at Meta to work on physics-based machine learning research!
  • [May 2022] Vincent is selected as a Givens Associate in the Mathmatics and Computer Science Division at Argonne National Labs, and will spend a summer there doing research on long-term traffic prediction.
  • [Jun 2020] Negin has accepted a Tenure-track Assistant Professor position in the Department of Engineering Systems and Environment at the University of Virginia.
  • [May 2020] Dr. Meidani has received a research grant to investigate optimal pooled testing strategies for COVID-19 community screening.
  • [Apr 2020] Negin's first author paper on fast surrogate-based voltage control for distribution energy systems have been published on IEEE Access.
  • [Feb 2020] Our team won the First Prize in the Data Challenge on Urban Travel Time, Speed, and Reliability at the 2020 TRB Annual Meeting. Congratulations to our graduate students Xiyue and Mohammad!
  • [May 2019] Mohammad Amin Nabian has been awarded the Dissertation Completion Fellowship from the Graduate College. Congratulations to Mohammad on this prestigious recognition.
  • [Apr 2019] Mohammad Amin Nabian has been selected by UIUC College of Engineering as one of the Mavis Future Faculty Fellows (MF3) for the 2019-2020 academic year. More on the fellowship program can be found here. Congrats to Mohammad!
  • [Mar 2019] Negin Alemazkoor has been selected as one of the Rising Stars in Computational and Data Sciences and will attend an event sponsored by the University of Texas at Austin and Sandia National Labs in April 2019. Congratulations to Negin!
  • [Nov 2018] Our group’s entry to the 2018 INFORMS Rail Problem Competition won the 2nd Place Prize. Congratulations to Mohammad and Negin. Full story.
  • [Aug 2018] Our journal paper on preconditioning for improved compressive sampling of polynomial surrogates is accepted for publication in Computational Methods in Applied Mechanics and Engineering. Congratulations to Negin.
  • [Jul 2018] Our conference paper on recursive data-driven prediction of traffic flow with faulty sensors is accepted for presentation at the 2018 IEEE International Conference on Intelligent Transportation Systems (IEEE ITSC 2018). Congratulations to Negin and Shiyu.
  • [Jul 2018] Our 2018 INFORMS Annual Meeting submission on efficient collection of connected vehicle data with precision guarantees is accepted. Congratulations to Negin.
  • [Feb 2018] Dr. Meidani receives NSF CAREER Award to study fast and accurate software tools for interdependent infrastructure systems. Full story.