Pendekatan Data Mining dalam Optimalisasi Margin Penjualan Adidas: Studi Klasterisasi dengan K-Means dan Fuzzy C-Means

Adidas is trying to stay competitive, especially when it comes to selling in the U.S. market. This study aims to cluster Adidas sales data in the US by comparing K-Means and Fuzzy C-Means clustering methods to achieve effective market segmentation. Data management is performed by first standardizing the data from the variables of units sold, total sales, operating profit, and profit margin. Cluster analysis is performed by determining the optimal cluster with the elbow plot, and the most optimal result is four clusters. Cluster performance evaluation was performed by comparing the Davies-Bouldin Index (DBI), Sum of Squares Within (SSW), and Sum of Squares Between (SSB) values of each method. The analysis results show that K-Means has a DBI of 0.98 and an SSW of 8645.74, while Fuzzy C-Means has a DBI of 0.983 and an SSW of 8678.49. Based on these results, K-Means is considered more optimal because it produces clusters that are more compact and separated. This finding can be the basis for developing a more targeted and efficient sales strategy according to the characteristics of each market segment.