Fuzzy Tsukamoto Untuk Merekomendasikan Pembelian Barang Berdasarkan Data Penjualan

The development of the sports industry, especially in futsal shoe sales, requires an inventory management strategy that is able to anticipate fluctuations in market demand. The main problem in this study is how to overcome the uncertainty of futsal shoe demand caused by variables such as trends, competition seasons, and changing consumer preferences. This study aims to develop a fuzzy logic-based purchase recommendation system using the Fuzzy Tsukamoto method to improve stock management efficiency. This study uses a quantitative approach with fuzzy method stages consisting of fuzzification, rule formation, inference, and defuzzification. The tools used are MATLAB software that supports the creation of fuzzy inference systems and graphic modeling. The study was conducted at the Pasifik Club Sports Sibolga Futsal Shoe Store by processing 1,098 historical sales data. The results of the study showed that the system built was able to recommend futsal shoe purchases with good accuracy, indicated by the Mean Absolute Percentage Error (MAPE) value of 15%. This system not only provides a popularity value for each shoe variant, but also helps stores avoid excess or shortage of stock. Thus, the Fuzzy Tsukamoto method is proven to be feasible to be used as a decision-making tool in retail inventory management.