<![CDATA[The increasing need for data-driven member management in the fitness center industry makes member segmentation important for identifying differences in activity, engagement, and financial contribution and supporting more targeted retention strategies. This study aims to group members of GYM JGM (Jaya Makmur Gym), Jember, based on Length, Recency, Frequency, and Monetary (LRFM) characteristics. The dataset used consists of 150 member data. The research process includes data preprocessing, LRFM feature selection, standardization using Z-Score, and the application of the K-Means algorithm. Determining the optimal number of clusters is done by comparing the Elbow Method, Davies-Bouldin Index (DBI), and Silhouette Score on candidate clusters k = 2 to k = 9. The evaluation results show that three clusters are the optimal configuration because they produce the lowest DBI value of 0.825 and the highest Silhouette Score of 0.430, and show a clear change point in the Elbow Method. The segmentation results produce three groups of members, namely Shorter-Term Moderate-Activity Members, High-Value Active Members, and Long-Term Low-Engagement Members. The segmentation shows that the K-Means algorithm based on LRFM features is able to identify differences in member characteristics and provide an analytical basis for developing retention strategies, increasing engagement, and reactivating GYM JGM members.]]>
Copyrights © 2026
