SETIAWAN, Edwin and Surarso, Bayu and Mutiara K.N., Dinar (2025) SEGMENTASI PELANGGAN BERDASARKAN ANALISIS RECENCY, FREQUENCY, MONETARY MENGGUNAKAN ALGORITMA K-MEANS PADA APPLE ECOSYSTEM. Masters thesis, UNIVERSITAS DIPONEGORO.

	Text 1-aCover.pdf Restricted to Repository staff only Download (1MB)
	Text 1-BAB I.pdf Download (282kB)
	Text 1-BAB II.pdf Download (577kB)
	Text 1-BAB III.pdf Restricted to Repository staff only Download (4MB)
	Text 1-BAB IV.pdf Restricted to Repository staff only Download (3MB)
	Text 1-BAB V.pdf Restricted to Repository staff only Download (275kB)
	Text 1-Daftar Pustaka.pdf Download (302kB)
	Text 1-Lampiran.pdf Restricted to Repository staff only Download (964kB)

Abstract

Salah satu perusahaan di Semarang yang bergerak di bidang jasa penjualan gadget memiliki sistem informasi Apple Ecosystem untuk penjualan produk dari brand eksklusif, Apple. Di dalamnya terdapat transaksi penjualan dan juga servis perangkat iPad, MacBook Air, MacBook Pro, AirPods, Mac, dan Apple Accessories. Penelitian ini menggunakan data transaksi pembelian dari pelanggan Apple Ecosystem periode tahun 2023. Metode analisis Recency, Frequency, Monetary (RFM) diterapkan untuk mengidentifikasi pola perilaku pelanggan, sedangkan algoritma K-Means digunakan untuk melakukan segmentasi pelanggan ke dalam klaster yang lebih spesifik. Untuk menentukan jumlah klaster optimal, penelitian ini menggunakan metode Elbow dan validasi Silhouette Coefficient. Hasil segmentasi menunjukkan tiga klaster utama: Cluster 2 (High Spenders) dengan 326 pelanggan, Cluster 0 (VIP Customers) dengan 473 pelanggan, dan Cluster 1 (Frequent Buyers) dengan 201 pelanggan. Selain itu, penelitian ini mengembangkan aplikasi web Customer Segmentation App (RFM Clustering) untuk mempermudah perusahaan dalam melakukan segmentasi pelanggan secara otomatis. Implementasi sistem ini memungkinkan perusahaan untuk mengoptimalkan strategi pemasaran berdasarkan data yang tersedia. Validasi menggunakan koefisien silhouette menghasilkan nilai 0.3524, yang menunjukkan bahwa segmentasi yang dihasilkan cukup baik. Dengan adanya penelitian ini, perusahaan dapat meningkatkan loyalitas pelanggan, mengefisiensikan biaya pemasaran, dan mengembangkan strategi bisnis berbasis data yang lebih optimal.
Keywords: Customer Segmentation, RFM Models, K-Means clustering, Elbow Method, Silhouette Coefficient

A company in Semarang engaged in the gadget sales service sector has an Apple Ecosystem information system for selling products from the exclusive brand, Apple. It includes sales transactions and servicing of iPad, MacBook Air, MacBook Pro, AirPods, Mac, and Apple Accessories. This research utilizes purchase transaction data from Apple Ecosystem customers for the 2023 period. Recency, Frequency, Monetary (RFM) analysis is applied to identify customer behavior patterns, while the K-Means algorithm is used to segment customers into more specific clusters. To determine the optimal number of clusters, this study uses the Elbow Method and Silhouette Coefficient validation. The segmentation results show three main clusters: Cluster 2 (High Spenders) with 326 customers, Cluster 0 (VIP Customers) with 473 customers, and Cluster 1 (Frequent Buyers) with 201 customers. Additionally, this study develops a web application called Customer Segmentation App (RFM Clustering) to facilitate companies in automatically segmenting customers. Implementing this system allows businesses to optimize marketing strategies based on available data. Validation using the silhouette coefficient resulted in a value of 0.3524, indicating that the segmentation produced is fairly good. With this research, companies can enhance customer loyalty, optimize marketing costs, and develop more data-driven business strategies.
Keywords: Customer Segmentation, RFM Models, K-Means clustering, Elbow Method, Silhouette Coefficient

Actions (login required)

View Item