IMPLEMENTATION OF FUZZY INVENTORY METHOD AND ARTIFICIAL NEURAL NETWORK IN DETERMINING SAFETY INVENTORY OF BAG PRODUCTS

Abstract

General Background: Effective inventory management is crucial for small and medium enterprises (SMEs) to address fluctuating demand and avoid shortages, especially in sectors like handcrafted products. Specific Background: In the context of PTK MSMEs (Karangtanjung Bag Craftsmen) in Sidoarjo Regency, bag product sales often vary monthly, necessitating accurate demand forecasting and optimal inventory levels. Knowledge Gap: While previous studies have explored demand prediction and inventory management, few have integrated advanced methodologies like Artificial Neural Networks (ANN) and Fuzzy Inventory approaches to cater specifically to SMEs in the handicraft sector. Aims: This research aims to predict the sales demand for bag products and establish safety inventory levels using ANN and Fuzzy Inventory methods, ultimately to control demand and reduce inventory costs. Results: The study yielded a Root Mean Square Error (RMSE) of 45.031 from the ANN analysis, indicating a good forecasting performance, while the Fuzzy Inventory method calculated a safety stock of 43,647 pieces for 2023. Novelty: The integration of ANN for demand forecasting and Fuzzy Inventory for safety stock determination offers a novel approach for SMEs, enabling them to respond proactively to market fluctuations. Implications: The findings provide a framework for MSMEs to enhance their inventory management practices, thus improving operational efficiency and reducing holding costs, which can significantly impact their sustainability and competitiveness in the market.