Implementation of Apriori Algorithm in Identifying Purchase Relationships at Bluder Cokro Pakuwon Mall

Bluder Cokro Store, located at Pakuwon Mall, specializes in traditional bluder bread with a wide range of flavor variations. This study aims to identify consumer purchasing patterns at the store to enhance promotional strategies and optimize product placement. The research applies the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology, which includes phases such as business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The dataset used consists of 4,371 transactions from October to December 2024. This study uses the Apriori algorithm to find patterns of association between products, with the goal of determining the scope of correlation between products and frequently co- purchased items. The results reveal nine significant association rules, with the strongest relationship observed between coklat keju and keju, having a support value of 0.100394 and a lift of 1.31. These findings indicate that strategic product placement and bundling promotions can enhance sales performance. Optimizing the store layout by placing coklat keju near coklat can increase purchase likelihood, while targeted discounts, such as "Buy coklat keju, get 10% off keju," can drive transaction values. This study serves as a recommendation framework rather than an experimental validation, offering insights on how transaction data and association rule mining can inform business decisions. The findings offer actionable insights for improving store layouts and promotional effectiveness, making this research valuable for retailers.