A Knowledge-Based Diagnostic Approach for the Launch of the Auto-Body Assembly Process

Abstract

This paper is the first attempt to implement a knowledge-based diagnostic approach for the auto-body assembly process launch. This approach enables quick detection and localization of assembly process faults based on in-line dimensional measurements. The proposed approach includes an auto-body assembly knowledge representation and a diagnostic reasoning mechanism. The knowledge representation is comprised of the product, tooling, process, and measurement representations in the form of hierarchical groups. The diagnostic reasoning performs fault diagnostic in three steps. First, an initial statistical analysis of measurement data is performed. Next, the Candidate Component and Candidate Station with the hypothetical fault are searched. Finally, the fault symptom is identified and the root cause is suggested. Two case studies are presented to demonstrate the implementation of the proposed method.