AI for Enhanced Oil Recovery
In response to the unpredictable nature of global energy markets, exacerbated by geopolitical shifts and supply chain disruptions, researchers at MIT World
Peace University (MIT-WPU) in Pune have pioneered sophisticated artificial intelligence (AI) and machine learning (ML) solutions. The primary objective is to elevate oil extraction from mature reservoirs and to generate more reliable production forecasts. A dedicated team from MIT-WPU's Petroleum Engineering Department, spearheaded by Professor Dr. Rajib Kumar Sinharay and his PhD student Dr. Hrishikesh K Chavan, has developed an ML model specifically designed to pinpoint the most effective Enhanced Oil Recovery (EOR) strategies for intricate reservoir environments. This model, meticulously trained on data gathered from numerous oil-producing sites worldwide, has demonstrated an impressive 91% accuracy in identifying optimal recovery techniques. This breakthrough technology drastically slashes the evaluation time for oil recovery methods from several months, as required by traditional approaches, to a mere handful of hours, thereby accelerating decision-making and operational efficiency in the field.
Forecasting and Rock Identification
Professor Samarth Patwardhan and his PhD candidate, Dr. Soumitra Nande, have also made significant strides by developing a deep learning model. This model possesses the remarkable capability to identify carbonate reservoir rocks with an exceptional 97% accuracy. These types of rock formations are particularly relevant as they are similar to those found in Bombay High, which is India's most substantial offshore oil field. The implications of this accurate rock identification are substantial for exploration and extraction strategies. The research conducted by Patwardhan and Nande, which was subsequently published in the esteemed Arabian Journal for Science and Engineering in 2025, underscores the growing importance of AI in understanding and optimizing subsurface resource extraction. Dr. Sinharay has emphasized the transformative potential of AI, stating that it can revolutionize reservoir management by providing data-driven tools for selecting the most efficient recovery methods and refining production forecasts, especially for fields that have been in operation for an extended period.
