University of Northumbria at Newcastle and Siemens Energy Industrial Turbomachinery Limited KTP 24_25 R3

Gas turbines contain thousands of components, each with a manufacturer and a lifespan, and when one part becomes obsolete the entire machine can be grounded. This project builds an AI-driven platform that predicts when specific turbine parts will go out of production, so operators can order replacements or redesign assemblies before a shortage hits. Currently, power plants and industrial sites that rely on gas turbines manage obsolescence reactively—scrambling for last-time buys or costly custom fabrications after a part disappears from the market. The platform uses machine learning to scan supply-chain data, component lifecycles, and manufacturing schedules, flagging at-risk parts months or years in advance. If it works, energy companies, petrochemical plants, and aviation operators could shift from crisis management to planned procurement. The immediate impact is on infrastructure reliability: fewer unplanned shutdowns, lower maintenance costs, and longer operational life for existing turbines. This is applied engineering research with a direct industrial payoff—no fundamental science claims, but a concrete tool for keeping critical machinery running.