JoWUA

Abstract

Cognitive manufacturing faces major challenges due to the increasing industrial competition. Cognitive manufacturing has been directed toward agile strategies that harness artificial intelligence, machine learning, and computer vision approaches to overcome the obstacles of four-generation manufacturing. The integration of these strategies represents an important research direction in the engineering research society. This paper develops a new Automated Defect Detection System (ADDS) utilizing the Echo State Wireless Sensor Network (ES-WSN) as a random shallow technique to detect defects in products through image analysis and training. The system employs artificial intelligence and real-world factory datasets to predict defects. Benchmarks, for instance, Accuracy, precision, recall, and F1, were used to evaluate the performance of the suggested model. The results indicate good performance compared with state-of-the-art methods. The accuracy detected at peak iteration (3000 iterations) is not less than 99.78 percent with an Average Relative Error (ARE) equal to 2.7 percent and a processing time equal to 80 milliseconds. ES-WSN improved product quality assurance to 98 percent. The research focused on using shallow learning as a computer vision technique on real-time images of specific products to develop an ADDS and improve product quality in cognitive manufacturing.