Tolerance estimation and metrology for reverse engineering based remanufacturing systems

Tolerances are critical to the product’s design, manufacturing, and quality. However, tolerances are often overlooked in a reverse engineering (RE) process for industrial applications, especially in legacy parts or spare parts remanufacturing. Ignoring tolerances could either unnecessarily call for high precision remanufacturing processes or make the reproduced parts unqualified. Additive manufacturing (AM) techniques are used in remanufacturing applications because of their ability to manufacturing intricated parts. Due to its layer-by-layer fabrication nature, the metrology for AM-created parts is drastically different when compared to traditional feature-generation processes. In this study, we first propose a novel way to classify manufacturing processes based on whether they directly identify or generate features, which could profoundly affect their metrology tools. Next, a systematic geometric inspection and tolerance estimation methodology for the RE system is proposed. A set of tools is developed to extract various geometric dimensional values from the point clouds based on their tolerancing types. Moreover, based on the domain knowledge in production process design and planning, methods are developed to estimate empirical tolerances from a small batch of legacy parts. Comparisons of empirical tolerances of real machined parts to their designed tolerances are presented to evaluate the performance of the proposed framework.