Detecting and Predicting Errors in Manufacturing Applications

David Goldstein

The "artificially intelligent" portions of real-world manufacturing systems must work in harmony with the rest of a parent system. Because artificial intelligence (AI) is considered a relatively high-risk technology, AI components should be responsible for their own correct behavior - regardless of constraints imposed by the embedding system. Therefore, AI subsystems must be able to at least predict or compensate for their own failures (if not for systems beyond their own scope.) This paper discusses a novel approach for predicting and compensating for errors. This paper specifically attempts to apply the approach to a manufacturing domain. We first describe the research we have already performed in previous papers as an introduction to our concerns. We then examine predictive computing and propose it in a form for addressing manufacturing faults. We conclude by describing the status of our work and the focus we intend to concentrate upon in future research.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.