Figure 9 AI-generated insights AI-generated insights from geospatial, meteorological, and historical leak rate data, used in conjunction with atmospheric dispersion models, can optimize sensor placement for maximum coverage and the timely detection of leak emissions.46 Automated generation Automatic generation of incident reports and the identification of available technicians and materials can expedite repair work. 1. Methane Detection and Monitoring Pilot ADNOC has launched a pilot project using methane detection technology for enhanced environmental management with high accuracy compared to the industry standard. The initiative uses passive FTIR spectroscopy, computer vision, and deep learning to monitor large areas (in combination with sensor networks) The technology provides real-time observation and remote operations. The system offers reliable alerts, 24/7 detection, and validation of methane, CO, CO2, TVOC, and H2S. It supports LIDAR efforts, identifies emission sources, and uses AI for plume modeling and heat mapping. 2. Real-Time Flare Combustion Monitoring ADNOC has implemented an AI-based solution for real-time flare combustion monitoring Using cost-effective CCTV cameras and deep learning, ADNOC assesses combustion efficiency (CE) with Temporal Standard Deviation (TSD) and flare event detection networks. The system provides reliable real-time data, supporting emission control strategies, aligning with environmental regulations, and enhancing sustainability by reducing emissions cost-effectively. Recent developments suggest that AI’s emergent abilities could support oil and gas players’ endeavors to measure, predict, and optimize complex systems to improve methane management. Potential opportunities include: Progress is being made. For example, Oxford University has developed an AI tool that scans geospatial data to detect leaks 20% more accurately than legacy tools.47 The model was trained on large volumes of data from NASA satellites. The researchers have made the base data and code open-source so that the tool is available to others. ADNOC is also developing and deploying a range of AI-based tools for managing methane that have shown early promise (see Figure 9). Methane detection Flaring management 3. Lab-Scale Flare Stack System ADNOC has developed a lab-scale flare stack system for studying and optimizing combustion, emissions, and safety protocols in industrial processes. This prototype simulates real-world conditions for systematic data collection and AI-driven flare management testing. Detailed monitoring and manipulation of combustion parameters generate high-quality data for AI models. This initiative can enhance the understanding of flare dynamics, reduce environmental impacts, and offer actionable insights for optimizing flare operations. ADNOC AI use cases for methane emissions and flaring reduction have shown early promise Algorithm analysis Algorithms that analyze data from internet of things (IoT) sensors can detect and even quantify leaks in near real time. 32 Powering Possible 31