Penerapan Metode Fuzzy Mamdani dan Fuzzy Sugeno dalam Penentuan Jumlah Produksi Produk Pakaian Konveksi Mutiara Naga

Abstract

Konveksi Mutiara Naga faces challenges in optimizing the production quantity of gamis (Muslim dresses), often resulting in either excess or shortage of stock. These issues arise from inaccurate production planning that fails to align with fluctuating market demand. This study aims to estimate the most optimal quantity of gamis production using a quantitative approach based on two years of historical data. The analysis was conducted using fuzzy logic methods, specifically Mamdani and Sugeno. Evaluation with the Symmetric Mean Absolute Percentage Error (SMAPE) shows that the Sugeno method yields better accuracy at 8.9%, compared to 9.09% from the Mamdani method. The optimal production quantity estimated by the Sugeno method is 351,028 units, while Mamdani estimates 359,449 units. These findings indicate that the Fuzzy Sugeno method is more effective and reliable in supporting production planning decisions, helping to minimize the risk of overstock or understock at Konveksi Mutiara Naga.