Pemodelan Segmentasi Transaksi Jual Beli Produk Menggunakan Pendekatan Model K-Means dan Subtractive Clustering Studi Kasus Survey Pada Beberapa Cabang Optik Retail

Abstract

A professional optical retail has several branches throughout Indonesia that offer various types of products ranging from eyeglass lenses, eyeglass frames, contact lenses and accessories from various brands. Sales data across all optical retail branches has a fairly large volume of 95,308 products during 2019. Where there are a variety of eyewear products that are quite diverse, making it difficult to classify the various types of variations for each type of eyewear product available. One way to group sales data based on customer characteristics can be using segmentation. To determine sales segmentation can be done by Clustering, namely by grouping data based on sales characteristics. The clustering process is carried out using the K-Means Algorithm and the data used is the result of a survey of sales of eyewear products at 3 optical branches. The results of calculations using the K-means algorithm in this study obtained as many as 3 clusters with different criteria for product brands, number of sales and prices in each branch. While the Subtractive calculation is looking for the maximum potential value available, so the number of clusters produced is 1 cluster which shows the best-selling eyewear products in each branch. The final result in the form of mapping customer behavior patterns can later be used as a strategy for providing products that will be sold in the following year by decision makers.