Rapid Damage Identification to Reduce Remanufacturing Costs
The objective of this project is to develop and validate a remanufacturability assessment method that will support decision making about the viability of remanufacturing a component. The proposed method is based on development of machine learning (ML) techniques for recognizing different types of component damage, embedding developed ML algorithms in low-cost, damage-identification hardware for use in-process at the remanufacturing factory floor, and using this in-process technique to develop a real- time estimate of remanufacturing costs for a component. Although most high-value, metal-alloy components can be remanufactured, sufficiently accurate and rapid decision making support tools are needed to significantly reduce remanufacturing costs and increase the throughput and volume of remanufactured components.