Minimarket Sales Optimization: Implementation of FP-Growth dan MongoDB With Python

Abstract

This study applies an integrated FP-Growth algorithm with MongoDB and Python to analyze 150,000 minimarket transaction records over a one-year period. The dataset includes transaction numbers, product names, quantities sold, transaction dates, purchase prices, and selling prices. The parameters of a minimum support of 0.001, a confidence of 0.01, and a lift above 1.0 are used to ensure relevant association rules. The analysis indicates that the discovered product association patterns can increase operational efficiency by up to 15%, particularly in instant food and ready-to-drink beverage categories. These data-driven strategies also boost sales volume by 12.3% and reduce dead stock by 8.7%. Beras MCS 5KG stands out as the most profitable product, with a margin of IDR 1,066,724,400. The main strength of this study lies in the integration of FP-Growth with MongoDB, enabling large-scale real-time analysis without generating candidate itemsets. This approach enhances data processing efficiency, allowing minimarkets to optimise inventory and promotional strategies more accurately.