IMPLEMENTATION OF DATA MINING FOR ANGKRINGAN SALES ANALYSIS USING THE APRIORI METHOD

Objective: This study aims to address the challenges faced by Angkringan Mdpl in analyzing sales data that affect stock management efficiency. The research seeks to identify purchasing patterns that can serve as a foundation for better inventory and marketing decisions. Method: The Apriori algorithm, a data mining technique, is employed to discover associations among sold items by calculating support and confidence values to generate valid association rules. The analysis uses transaction data from June and July 2024, with a minimum support threshold of 2% and a minimum confidence level of 5%. Results: The testing process produced five pairs of item combinations with strong and valid association rules, as confirmed by their lift values. These findings enable Angkringan Mdpl to enhance stock control by prioritizing frequently purchased products and recommending complementary items effectively. Novelty: This study provides a data-driven approach to micro-scale food business management by applying the Apriori algorithm to optimize stock planning and sales strategies in small local enterprises, demonstrating the algorithm’s practical value beyond large-scale retail contexts.