3 Defect Root Cause Analysis in Production Lines Based on Hierarchical Baysian Network

Quality control is of crucial importance in many production lines such as car manufacturing. It shall enable detection and rejection any defective product and prevent its delivery to the market. In order to prevent high losses due to defective products, some measures must be taken. In this paper a new approach is presented that enables an effective defect root cause analysis (DRCA) in production lines for quality improvements in manufacturing system and process-engineering departments. This method includes the development of a knowledge model, based on information obtained from experience, various manufacturing data source and defect related knowledge. Due to the increasing level of complexity in production lines and various uncertainties in system parameters a hierarchical Baysian network is used. which identifies the most probable root cause of the defect. This method has been applied to the defective vehicle body surfaces in paint shop of an automotive assembly plant. Both simulation results and practical tests demonstrate the acceptable performance of the proposed method.