Application of Support Vector Machines to Fault Diagnosis and Automated Repair

Download

PDF Download (137Kb)

Description/Abstract

In this paper we consider the benefits of applying modern machine learning techniques to the problem of Fault Diagnosis and Automated Repair. In the modern manufacturing environment, many aspects of the production line are logged automatically by various systems. These records are put to a multitude of uses including assisting stock control, and monitoring and improving the manufacturing process. This approach has lead to the accumulation of a huge amount of high-dimensional data, and does require new methods to handle it. In this paper we ask if the information commonly held by many companies can be used to assist the repair of faulty products on the production line. We examine the possibility of using pattern recognition techniques to determine correct repairs for faults from past production history. The relative merits of this method compared to other approaches (such as model-based reasoning) are also discussed. Finally, we give some preliminary results which indicate that pattern recognition methods such as the highly acclaimed Support Vector machine can be successfully applied in this area.