Multilevel Association Rules on Customers’ Buying Pattern Based on Sales Transactions: A Study Case in Retail

Abstract

The retail industry is a vital sector of the world economy and is characterized by fierce competition, tight profit margins, and demanding consumers. Understanding customer buying behavior patterns is essential in devising the best retail strategy to enhance product sales. This research aims to comprehend customer shopping behaviors based on retail sales transactions and formulate the best strategies. By employing multi-level association rules, the dataset is arranged hierarchically into categories, sub-categories, and items. The sales transaction data used comprises 5830 transaction records over a month. The results of this study reveal 24 associations of categories, 49 associations of sub-categories, and 12 associations of product items. Moreover, the proposed marketing strategy offers recommendations including store layout improvement, planogram design, and bundled product offerings. This research addresses the gap in empirical evidence from a previous study and suggests further observation from diverse locations to authenticate the findings, which may yield various outcomes

Keywords

• buying behaviour
• data mining
• fp-growth
• marketing logistics
• retail strategies.