Abstract
Defect detection and classification are important tasks for both product quality assurance and process improvement in the manufacturing industry. In leadframe manufacturing, samples of cut leadframes are taken immediately after stamping and examined by human inspectors. Early defect detection is critical for reducing waste of raw material and achieving a high quality product. In this paper, the performance of several participants was tested on the basis of inspection speed and accuracy. The tests were carried out using images of three leadframe designs with varying number/type of stamping defects. Response surface methodology was used to analyze the test results. In addition to designing better training systems, the results can be used by manufacturers to monitor their inspectors' performance. Due to the generic nature of the inspection task, the results can be further applied in other industries as well.
Get full access to this article
View all access options for this article.