Tata Steel And IIT Bombay Team Receives INFORMS Franz Edelman Finalist Award 2024

Education Edited by
Tata Steel And IIT Bombay Team Receives INFORMS Franz Edelman Finalist Award 2024

Tata Steel And IIT Bombay Team Receives INFORMS Franz Edelman Finalist Award 2024

Tata Steel, India, and IIT Bombay team received the INFORMS Franz Edelman Finalist Award 2024 during the Edelman Gala night at INFORMS Business Analytics Conference 2024, Orlando, USA. The award recognized the novel Model Predictive Control (MPC) technology-based solution developed in-house by Tata Steel in consultation with the IIT Bombay, for the continuous annealing furnace.

INFORMS Franz Edelman Finalist Award is the world’s leading Operations Research and Analytics Award that is given by the Institute for Operations Research and the Management Sciences (INFORMS Society, USA). The award is conferred each year for successful industrial applications of operations research. Many other teams including American Airlines, McDonald’s China, ALDI Sud Germany, Molslinjen Denmark, and Transvision Netherlands reached the finals after two rounds of review scrutiny spanning four months by INFORMS Society judges.

Tata Steel and IIT Bombay were inducted as members of the prestigious Franz Edelman Academy. Since 1972, membership in the Academy represents an extraordinary contribution to society through the innovative application of analytical decision-making in business. Franz Edelman Academy members serve as role models for other organizations and exemplify that challenges can be faced successfully and opportunities for improvement can be found in the very organization through the innovative application of analytics.

Model Predictive Control (MPC) technology-based solution dynamics

In the steel industry, a continuous annealing furnace is crucial for controlling the quality of cold-rolled steel strips through heat treatment. Achieving the optimal target strip temperature during the annealing process is vital as it determines the microstructure and mechanical properties of the product coil.

However, controlling this process is challenging due to slow furnace temperature dynamics, significant time delays, frequent changes in the steel mass flow rate, the target annealing temperature changes induced by steel grade transitions, and multivariable interactions within the furnace zones. To address these challenges, a novel Model Predictive Control (MPC) technology-based solution has been developed in-house by Tata Steel, India for the continuous annealing furnace with the Indian Institute of Technology Bombay as the academic consultant.

The system is controlled by repeatedly solving a sequence of finite horizon optimal control problems online, once every minute. This approach ensures smooth changes in strip temperature profile during steel grade transitions and prevents the violation of operating constraints.

Benefits of the MPC-based technology solution

The MPC-based supervisory optimal control solution has been successfully implemented in real-time operations since January 2023. The systematic implementation of the optimal control solution has led to a significant improvement in the proportion of annealed products meeting the premium quality band and prevented the redirection of 13,000 tons of material outside the control band annually.

This solution has reduced fuel consumption per ton of steel by 8%, which has resulted in an 8% reduction in CO2 emissions annually. This improved control and fuel savings have translated to significant financial benefits. Furthermore, MPC implementation has resulted in a 20% improvement in premium product quality, helping retain customers and acquire new business in the highly competitive automotive steel market.

Tata Steel India and IIT Bombay team Anil Pujari, Chief of Flat Products Manufacturing, Dr. Jose Korath, Chief of Intelligent Systems & Mathematical Modelling group, Sujit Jaganade, Lead Engineer, and Prof. Sachin Patwardhan were present at the award function with INFORMS Executive Director Dr. Elena Gerstmann and Prof Julie Swann, (University of North Carolina and INFORMS President 2024).