Skip to main content
Alinson Santos Xavier preview image

Alinson Santos Xavier

Principal Computational Scientist

Alinson Santos Xavier is a computational scientist advancing the efficiency and reliability of the U.S. power grid through cutting-edge research in AI, mathematical optimization and high-performance computing.

Biography

Highlights

Alinson Santos Xavier is a principal computational scientist at Argonne National Laboratory, developing advanced methods that combine optimization, machine learning, and high-performance computing to improve power system operations and planning. He leads several DOE-funded projects and is the developer of open-source tools like MIPLearn and UnitCommitment.jl, which support real-world grid operations and energy market analysis. His work bridges innovation in computation with practical applications across the energy sector.

Research Focus

Xavier’s research lies at the intersection of artificial intelligence, mathematical optimization, and high-performance computing, with the goal of solving complex problems in power systems and critical infrastructure. He focuses on developing learning-enhanced optimization algorithms that significantly accelerate the solution of large-scale problems, such as those found in grid operations and energy market modeling. His work includes creating tools for supply chain optimization, grid situational awareness, and reverse logistics, as well as advancing hybrid AI-optimization approaches to improve both computational performance and decision transparency. 

Impact

Xavier’s work is helping to transform how energy infrastructure is modeled, optimized, and secured. By developing hybrid approaches that integrate AI with traditional optimization methods, he is enabling faster, more accurate solutions to the complex problems facing modern power grids. His research has demonstrated 10x acceleration in critical optimization processes without sacrificing solution quality or interpretability, an essential requirement for grid operators.  

These advancements allow stakeholders to make faster, better-informed decisions about energy distribution, planning, and resilience. His open-source software is already in use by national laboratories, universities, and industry partners, bridging the gap between academic research and real-world implementation. 

Alinson Santos Xavier was always drawn to computers and problem-solving. His early interest in computing began with his first personal computer and evolved into an academic path that blended computer science, mathematics, and real-world applications. After earning his bachelor’s and master’s degrees in Computer Science from Universidade Federal do Ceará in Brazil, he pursued a PhD in Mathematics (Combinatorics and Optimization) at the University of Waterloo in Canada.

At Argonne, Xavier leads research to address some of the most computationally demanding challenges in power systems. He has pioneered learning-enhanced optimization methods that integrate AI into traditional solvers used by grid operators, improving speed without compromising critical standards like feasibility, optimality, and transparency. This hybrid approach is essential in a sector where decision-making must be both fast and reliable.

His open-source software tools, including MIPLearn and UnitCommitment.jl, are publicly available and designed to be adopted and extended by the broader research community. These tools represent a foundational shift in how optimization problems are approached and used in real-world settings, including collaborations with grid operators such as MISO and research institutions like Georgia Tech.

Xavier’s work extends beyond power grid optimization. He has developed methods for supply chain modeling in reverse logistics, natural gas pipeline monitoring using natural language processing (NLP), and secure outsourcing of grid computations to cloud platforms. Each project reflects his commitment to creating innovative and practical solutions.

In addition to his scientific contributions, Xavier advocates for openness and collaboration in research. By developing accessible, high-performance tools, he ensures that others can build on his work and that insights are shared across institutional boundaries.

Outside work, Xavier has created an enormously popular habit-tracking app, the Loop Habit Tracker, with over 5 million downloads. He also enjoys playing the guitar and spending time with his young family. His dedication to real-world impact and user-centered design carries through both his professional and personal projects. 

Read the latest Argonne news about Alinson Santos Xavier:
Outage Prediction and Grid Vulnerability Identification Using Machine Learning on Utility Outage Data

This tool is not only useful for disaster response, but also provides a quantitative evaluation framework for system resilience against certain types of extreme weather events.

A General Framework for AI-Accelerated Power Systems Optimization

This general framework for the next-generation of power system optimization tools uses Artificial Intelligence (AI) to automatically identify patterns in previously-solved optimization subproblems and effectively uses this information to accelerate the solution of new ones. 

Education

  • Ph.D. in Combinatorics & Optimization — University of Waterloo (2017)
  • M.Sc. in Computer Science — Federal University of Ceará, Brazil (2011)
  • B. Sc. in Computer Science — Federal University of Ceará, Brazil (2018) 

Honors and Awards

  • Impact Argonne Award (2020)
  • Outstanding Reviewer, IEEE Transactions on Power Systems (2018)
  • Honorable Mention, ACM South American Programming Contest (2008)
  • Honorable Mention, ACM South American Programming Contest (2007)
  • Silver Medal, Brazilian Olympiad in Informatics, Brazilian Computing Society (2005)
  • Silver Medal, Brazilian Olympiad in Informatics, Brazilian Computing Society (2004)
  • Silver Medal, Brazilian Olympiad in Informatics, Brazilian Computing Society (2003) 

Select Publications

Theses and Dissertation

Open-Source Software: