Evaluating the Quality of Manufactured Products by Providing an Approach based on the ANFIS Neural-Fuzzy Network (Case Study: Khazar Plastic Manufacturing and Industrial Company)

Currently, the need to expand accurate and rapid quality assurance to provide high quality and safe manufactured products; Yes, quality focuses more on internal issues that are used to control and improve internal processes and seek to improve performance for customer satisfaction and competitiveness. The quality of products and processes is emphasized in industry, government and society. The quality performance evaluation system is highly dependent on the identification and selection of critical success factors and its indicators in the framework of quality management. To achieve this goal, there are problems in the production industry, such as low production efficiency, low accuracy and lack of innovation in products.

Based on this, a plan has been proposed to introduce the artificial intelligence method of manufacturing companies to solve the above problems and, as a result, improve product quality and production efficiency. For this purpose, a network based on an adaptive neuro-fuzzy inference system is presented to evaluate the accuracy of the results and compare its efficiency. The proposed method is opposed to hard calculations and will save time and money. In industries, the volume of data has increased, which has led to the emergence of new concepts such as big data analytics, and the limitations and advantages of artificial intelligence-based solutions are discussed to creative attention to new solutions and new directions in manufacturing, commercial and service industries to improve efficiency. Stimulate your processes, increase the value of your solutions, and design new products to find new businesses and markets. Almost all international roadmaps focused on innovation and research include artificial intelligence as a fundamental driver of future technology. A challenge for traditional methods in quality management is the management of high-dimensional and non-linear production data. To solve these problems, a process based on artificial intelligence has been developed in this research to improve product quality and production efficiency. In this research, an adaptive neural-fuzzy inference model that has the advantages of neural network and fuzzy inference together, to evaluate and extracts the quality level of manufactured products; And it is proposed to infer how a set of production parameters and process quality of a production system are related.

In order to train the proposed model, the data related to the quality process of one piece from the production line of Khazar Plastic Industrial Company was used and 550 data related to the quality process with emphasis on influential variables including "appearance standard, external diameter of large gear, thickness of gear, The length of the metal shaft, the height of the metal shaft, the external diameter of the metal shaft" are considered as input variables and final quality as output variables. Finally, the accuracy of the results and the effectiveness of the proposed adaptive neuro-fuzzy inference model have been evaluated using statistical indices of correlation coefficient and root-mean-square error. 

the results show the data used to evaluate the quality of the production part in the adaptive-proposal neuro-fuzzy method; With a correlation coefficient of 0.95 and a mean square error of 0.42869, it provided a good match between the quality of the model output and the actual values.