Integration of Data Driven Approach Using AHP, Clustering, AR-MBA, and Classification to Improve Performance of MSME Restaurant

Abstract

The restaurant industry in Indonesia has experienced rapid growth alongside technological advancements and digitalization. Innovations in service, marketing, and operations have become key factors in improving business performance amid intense competition. SMEs such as Warung Makan Mantap Jaya in Yogyakarta need to implement the right strategies to adapt, enhance customer satisfaction, and strengthen their competitiveness. This study aims to formulate data-driven performance improvement strategies by integrating the Analytical Hierarchy Process (AHP), Clustering, Association Rule–Market Basket Analysis (AR-MBA), and Classification methods using data from 150 respondents. The results indicate that, based on the AHP method, renovating the premises is the top priority with the highest weight of 0.389, followed by discount innovations (0.365) and menu promotions (0.235). Clustering analysis identified three customer segments with distinct characteristics, where the majority of customers fall into the dissatisfied group, characterized by low visit frequency, low spending, and dissatisfaction with service. Meanwhile, some customers exhibit high satisfaction levels with higher visit frequency and spending. AR-MBA analysis identified purchasing patterns that generated menu bundling recommendations, such as combinations of Iced Tea, Chicken with Water Spinach, and Bakwan; Fried Egg and Mineral Water; and Soy Sauce Chicken and Tea Juice. Additionally, a classification method was used to predict customer satisfaction levels based on the resulting data patterns