Akademik Veri Yönetim Sistemi
NEURAL COMPUTING AND APPLICATIONS, cilt.1, ss.1-20, 2024 (Scopus)
Companies carry out the control process, including quality, with similar parameters, regardless of product type. Product-based classification must be made in final/completed product control, and the improving parameters must also be determined. Sending faulty products to the end user is an external failure cost and difficult to compensate. This study aims to design a new control process to prevent companies from sending faulty products to customers. First, criteria to improve product quality indicators were determined, and criterion weights were found using the analytic hierarchy process. Then, the Technique for Order Preference by Similarity to Ideal was used to rank the products. Finally, according to the importance levels of the products, the type of sampling plan, inspection levels, sample size, and acceptance numbers were determined using the acceptance sampling method. The most important criterion for the company is quality, with approximately 60% weight, followed by cost, with 20% weight, and demand and product complexity, with equal weight of 10% each. When we compared the existing and new control numbers, it was seen that 14% of the products had a higher control number than the new control number, and 80% needed to increase the control number above 100%. In the remaining 6%, it was determined that the existing and new control numbers were the same.