A major Asian oil & gas company wanted to improve the efficiency of its upstream operations by increasing the reliability of the Gas Turbine Compressors (GTC) used across its fields.
GTCs are equipped with hundreds of sensors that constantly monitor the state of this critical equipment using predefined thresholds to maintain production. However, the system
monitoring the GTCs generated thousands of alerts every month for each compressor unit and could not prioritize alerts or provide prescriptive insights. As a result, engineers manually sifted through alarms to identify anomalies, perform failure mode analyses, and recalibrate thresholds. This time-intensive process required a high level of expertise and knowledge of compressor systems. To address these challenges and increase the reliability of its compressors, the oil & gas company developed a prototype machine learning model to detect anomalous equipment behaviors with greater accuracy. The company then selected Baker Hughes and C3 AI® to deploy the model at scale using the BHC3TM Reliability application.