Improving Solver Run-time Performance - OR Tutorials for Professionals

The purpose of this Tutorials series (cum training) is to cover topics which are not typically covered in academics and are quite relevant for an OR practitioner. Each talk/lecture will be given by a seasoned OR practitioner teaching one topic in-depth. The focus will be on learning practical tips and techniques (keeping the theory to a minimum).

Topic: Improving Solver Run-time Performance

Abstract: Commercial Mixed Integer Programming (MIP) solvers have become increasingly powerful, and capable of solving complex optimization problems with impressive efficiency over the years. However, they often face challenges in scaling effectively due to various factors. In this talk, we will explore key aspects that affect solver performance, starting with analysing solver node logs to pinpoint specific causes of slow performance and better using solver features to improve efficiency. We will also discuss how to identify weak MIP formulations, as recognizing and addressing weaknesses in a formulation is crucial for enhancing solver performance and achieving better results.

Bio: Dr. Alok Patel is the founder of DecisionOpt, a consulting firm specializing in advanced analytics and AI-driven solutions for complex decision-making challenges. With over a decade of industry experience, he has applied operations research and machine learning techniques to optimize supply chains, logistics, and large-scale decision processes. He holds a Master’s and Ph.D. in Operations Research from IEOR, IIT Bombay. Dr. Patel is also the creator of ORGuru, a thriving LinkedIn community with over 2,000 members, uniting OR professionals, academicians, and students to foster collaboration and learning. His expertise spans optimization, simulation, and data science, with a proven track record of delivering impactful solutions to real-world problems. Learn more about him here.