<![CDATA[A mask manufacturer in Indonesia with a production of 4 million masks per day for various types of masks. However, in the production process there are still many defective and unsalable masks that enter the stock of goods to be sent, this is due to the quality control process that is still manual. So that to reduce product defects, it is necessary to mitigate by creating a system that can detect defective products, to facilitate the quality control process, an intelligent computing system is needed so that it is expected to reduce mask production defects to build this computational model will be carried out in several stages. The first stage will be a field study to obtain samples of defective and perfect products. The second stage builds a computational model, this model is built based on the Convolution Neural Network (CNN) method and the third stage builds a system that suits the needs in the field and tests the system against the company's needs. The purpose of this research is to produce a good and perfect defective product detection system so that it can be useful for reducing defective products that pass the quality control stage. From this research, if the process is run by entering existing data, it produces an accuracy percentage of 99% of the 750 data tested. While in real time testing, a percentage of 96.4% was obtained using 28 data.]]>
Copyrights © 2023
