Implementation of K-Means Clustering Algorithm in Food Sales (Case Study: Ayam Betutu at Warung Wardana)

Abstract

Ayam Betutu Warung Wardana is one of
the small food and beverage businesses which is in great demand to relax or to do work. To maximize
this business, managers must know the needs of customers who have visited, therefore they can
improve services according to what is needed. Besides that, there are problems that often occur,
namely problems regarding the menu that has the most potential to be improved and the menu that is
most in demand by consumers. From the problems that have been described above, this study applies
the K-means algorithm clustering method to determine consumer interest in a food and beverage
menu at Ayam Betutu Warung Wardana. In this study, the authors used a framework based on the
Cross Industry Standard Process for Data Mining (CRISP-DM) method. This study produced the best
two clusters with the Silhoutte and DBI validation methods. Testing using the Silhoutte Coefficient
shows a value of 0.44 and it is the best value close to the Davies Bouldin Index (DBI) showing the
value with the smallest ratio at 0.9030707. This is good because the ratio is close to zero, the better
the K-Means grouping.