2 minute readPuneMarch 24, 2026 10:29 PM IST
As geopolitical tensions and oil supply chain disruptions destabilize global energy markets, researchers at the Massachusetts Institute of Technology World Peace University (MIT-WPU) in Pune have developed advanced artificial intelligence (AI) and machine learning (ML) models to improve oil recovery from mature reservoirs and more accurately predict production.
Researchers in the MIT-WPU School of Petroleum Engineering are applying AI to tackle complex challenges in petroleum reservoir management. A team led by Dr. Rajiv Kumar Sinhalai, a professor in the department, along with PhD student Dr. Hrishikesh K. Chavan, developed a machine learning model that can identify the optimal enhanced oil recovery (EOR) technique for complex oil reservoirs.
The model was trained using data from numerous oil producing fields around the world and achieved 91% accuracy in predicting the most effective recovery method. The research results were published in the international journal Petroleum Science and Technology. AI-based models significantly reduce the time required to evaluate oil recovery strategies from months to just hours with traditional methods.
“Artificial intelligence has the potential to transform reservoir management in the oil and gas industry,” Dr. Sinhalai said in a statement. “Our research focuses on developing data-driven tools to help operators select the most effective recovery techniques and develop more accurate production forecasts, especially in mature fields.”
In another development, Professor Samarth Patwardhan and PhD student Dr Soumitra Nande have developed a deep learning model that can identify carbonate rocks with 97 per cent accuracy. These formations are similar to those found at Bombay High, India’s largest offshore oil field. Their research was published in the Arabian Journal for Science and Engineering in 2025.
