The Inspection Of Design Defects In Electronic Products

With the development of technology, the electronic products are developing in a trend towards being integrated, systematic and intelligent. So the electronic product design complexity increases sharply. All of this leads to the rising risk of the design flaws. Nowadays, research achievements about electronic product design defects are very little. What had been achieved mainly focus on the specific product defect recognition, recognition tools development, or from the process management and quality management perspectives to explore the design flaws in the project. But research about assessment of the severity of the design flaws, defects distribution probability calculation and defects factors diagnosis in the development of product are still few. This paper starts with the study of electronic product design defect formation mechanism and the classification and identification of design flaws and their factors, and then studies how to use rough sets and Bayesian network theory to solve those problems.First, this paper analyses the electronic product design defect formation mechanism from the perspectives of process management and the dynamic changes of the influencing factors. Then, the paper discusses the classification methods of design flaws from different perspectives, analyses and summarizes the influencing factors of design flaws, defect location, and defect recognition methods. The paper connects the research work with the defect tracking system, designs defect tracking record based on the contents mentioned previously and applies it to the company’s design defect management.This paper represents a design defects inspection method based on rough sets and Bayesian network theory. The method, firstly, extracts defect tracking records of the completed project from the defect tracking system, processing those records by the rough sets decision table, and uses them for the construction of Bayesian network inference model. Then, the structure and parameters of the Bayesian network inference model are adjusted according to the current similar project. Finally, the paper applies the Bayesian network inference model to the current similar project to access the design defect severity, calculate the defect distribute probability and diagnose the design defect’s influencing factors.In the end of this paper, the design defect inspection method is applied to the development of the voltage monitoring system. It is proved that the method is feasible and effective.