Design of Web-Based E-Smart Cashier using Apriori for Purchase Patterns and Product Recommendations

A web-based E-Smart Cashier was developed to integrate product and category management, stock-in purchases, sales transactions, transaction history, sales reporting, user management, system configuration, purchasing-pattern analysis, and product recommendation. Rapid Application Development was applied through requirements planning, user design, construction, and cutover. The Apriori algorithm processed transaction details using minimum support and minimum confidence to generate association rules. Functional black-box testing covered 13 scenarios and all scenarios were valid, producing a 100% functional success rate within the tested scope. Apriori testing for 1-18 August 2026 used 10% minimum support and 30% minimum confidence, analyzed two transactions, and generated six rules. Every rule had 50% support; four rules reached 100% confidence and two rules reached 50%. A sales report for 1-6 August 2026 contained three paid transactions totaling Rp21,500. The results demonstrate the integration of cashier operations with transaction-based association analysis, while the small Apriori dataset limits generalization of the recommendation patterns. Keywords: Apriori algorithm; E-Smart Cashier; Market basket analysis; Point of sale; Product recommendation