A two-phased pilot was conducted by Occidental to test the viability of AI vision at an oil processing facility construction site in the Permian Basin. A commercial platform was selected because of its robust suite of machine learning algorithms designed to identify safety hazards from a video stream. Additional features included analytics on the severity, type, and location of the hazards. Cameras were positioned at the edge of the location, facing areas of dense activity that included lifting, welding, trenching, driving, electrical work, and mechanical work. The first phase proved that AI vision was able to identify hazards with a high degree of accuracy. Observations were reported through an online dashboard complete with categorization, with still images and video clips. Data analytics provided insights that Occidental was able to take back to operations personnel. The second phase explored solutions for deploying AI vision where the communications infrastructure is limited.
πHighlights:π Results from an application on a commercial platform in the Permian Basin, which had a robust suite of machine learning algorithms designed to identify safety hazards.π The first phase proved that AI vision was able to identify hazards with a high degree of accuracyπ The second phase explored deploying AI vision where the communications infrastructure is limited. π’ The July 2025 hashtag#MIT study - https://lnkd.in/gk_JQQet: 95% of organizations are getting ZERO return from their GenAI investments. And, the 5% are extracting HUGE value! How we use the AI technology makes a difference β‘ποΈ Kyle Richter, PE, PMP , a registered professional engineer in the State of Texas and a certified Project Manager, was the 2018 Chairman of the SPE Permian Basin Section. At Oxy, he works on CO2 flood optimization and other sustainability projects. He is an experienced Operations and infrastructure projects leader.