Melola Discovery

← Melola Discovery

Open access Jurnal SINTA 3

Perbandingan Clustering Berbasis RFM dan Implementasi K-Means untuk Segementasi Pelanggan Bisnis Laundry

SemanTIK: Teknik Informasi · Vol. 12 No. 1 (2026) · pp. 22–32

DOI
10.55679/semantik.v12i1.275
ISSN
2502-8928, 2460-1446
Published
2026-04-26
Publisher
Informatics Engineering Department of Halu Oleo University
Language
eng
License
creativecommons.org/licenses/by-sa/4.0

Abstract

Bisnis laundry seperti Anty Laundry masih mengelola program loyalitas secara manual hanya mengandalkan frekuensi transaksi untuk menentukan pelanggan prioritas tanpa mempertimbangkan kebaruan dan nilai transaksi sehingga strategi pemasaran menjadi tidak tepat sasaran dan potensi retensi pelanggan bernilai tinggi tidak optimal. Penelitian ini membandingkan tiga metode clustering (K-Means, K-Medoids, dan Hierarchical Clustering) untuk segmentasi pelanggan prioritas berbasis analisis Recency, Frequency, dan Monetary (RFM), sekaligus mengimplementasikan metode terbaik dalam sistem informasi berbasis web. Data penelitian terdiri dari 2.549 transaksi valid dari 203 pelanggan unik periode Oktober 2024–Oktober 2025. StandardScaler digunakan untuk normalisasi data dan metode elbow menentukan jumlah cluster optimal (k=5). Evaluasi menggunakan Silhouette Score, Davies-Bouldin Index, dan Calinski-Harabasz Index menunjukkan K-Means mencapai hasil terbaik dengan nilai 0.5440, 0.5005, dan 282.18, mengungguli K-Medoids (0.4790, 0.7034, 145.52) dan Hierarchical Clustering (0.5141, 0.5239, 251.73). Lima segmen pelanggan teridentifikasi: Inactive Customer (36.95%), Regular Customer (49.75%), High Value Customer (11.82%), VIP Customer (0.99%), dan Top Spender (0.49%). K-Means diimplementasikan menggunakan Streamlit dengan segmentasi otomatis dan kemampuan ekspor untuk mendukung strategi pemasaran tepat sasaran per segmen. Laundry businesses such as Anty Laundry still manage loyalty programs manually — relying solely on transaction frequency to determine priority customers without considering recency and monetary value — resulting in poorly targeted marketing strategies and suboptimal retention of high-value customers. This study compares three clustering methods (K-Means, K-Medoids, and Hierarchical Clustering) for priority customer segmentation based on Recency, Frequency, and Monetary (RFM) analysis, while implementing the best-performing method in a web-based information system. The dataset consisted of 2,549 valid transactions from 203 unique customers covering October 2024 to October 2025. StandardScaler was applied for data normalization and the elbow method determined the optimal cluster number (k=5). Evaluation using Silhouette Score, Davies-Bouldin Index, and Calinski-Harabasz Index showed K-Means achieved the best results with values of 0.5440, 0.5005, and 282.18, outperforming K-Medoids (0.4790, 0.7034, 145.52) and Hierarchical Clustering (0.5141, 0.5239, 251.73). Five customer segments were identified: Inactive Customers (36.95%), Regular Customers (49.75%), High Value Customers (11.82%), VIP Customers (0.99%), and Top Spenders (0.49%). K-Means was implemented using Streamlit with automatic segmentation and export capabilities to support targeted marketing strategies for each segment.

Keywords

Collaboration network of the authors

7 authors · 9 co-authorship links

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • Manado State University
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
G D P MaramisManado State University1
Mesiasi Anjelika SupitManado State University1
Tesalonika Natalia RondonuwuManado State University1
Irene Realyta Halldy Trosi TangkawarowManado State University1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://semantik.uho.ac.id/index.php/journal/oai, first harvested 26 Okt 2026, last seen 26 Okt 2026.
Original metadata (oai_dc) 
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Perbandingan Clustering Berbasis RFM dan Implementasi K-Means untuk Segementasi Pelanggan Bisnis Laundry</dc:title>
	<dc:creator>Zai, Nibertin</dc:creator>
	<dc:creator>Kenap, Audy Aldrin</dc:creator>
	<dc:creator>Hasibuan, Alfiansyah</dc:creator>
	<dc:subject xml:lang="en-US">analysis RFM</dc:subject>
	<dc:subject xml:lang="en-US">Pengelompokan K-Means</dc:subject>
	<dc:subject xml:lang="en-US">Segmentasi Pelanggan</dc:subject>
	<dc:subject xml:lang="en-US">Aplikasi Streamlit</dc:subject>
	<dc:subject xml:lang="en-US">Strategi Pemasaran</dc:subject>
	<dc:description xml:lang="en-US">Bisnis laundry seperti Anty Laundry masih mengelola program loyalitas secara manual hanya mengandalkan frekuensi transaksi untuk menentukan pelanggan prioritas tanpa mempertimbangkan kebaruan dan nilai transaksi sehingga strategi pemasaran menjadi tidak tepat sasaran dan potensi retensi pelanggan bernilai tinggi tidak optimal. Penelitian ini membandingkan tiga metode clustering (K-Means, K-Medoids, dan Hierarchical Clustering) untuk segmentasi pelanggan prioritas berbasis analisis Recency, Frequency, dan Monetary (RFM), sekaligus mengimplementasikan metode terbaik dalam sistem informasi berbasis web. Data penelitian terdiri dari 2.549 transaksi valid dari 203 pelanggan unik periode Oktober 2024–Oktober 2025. StandardScaler digunakan untuk normalisasi data dan metode elbow menentukan jumlah cluster optimal (k=5). Evaluasi menggunakan Silhouette Score, Davies-Bouldin Index, dan Calinski-Harabasz Index menunjukkan K-Means mencapai hasil terbaik dengan nilai 0.5440, 0.5005, dan 282.18, mengungguli K-Medoids (0.4790, 0.7034, 145.52) dan Hierarchical Clustering (0.5141, 0.5239, 251.73). Lima segmen pelanggan teridentifikasi: Inactive Customer (36.95%), Regular Customer (49.75%), High Value Customer (11.82%), VIP Customer (0.99%), dan Top Spender (0.49%). K-Means diimplementasikan menggunakan Streamlit dengan segmentasi otomatis dan kemampuan ekspor untuk mendukung strategi pemasaran tepat sasaran per segmen.
Laundry businesses such as Anty Laundry still manage loyalty programs manually — relying solely on transaction frequency to determine priority customers without considering recency and monetary value — resulting in poorly targeted marketing strategies and suboptimal retention of high-value customers. This study compares three clustering methods (K-Means, K-Medoids, and Hierarchical Clustering) for priority customer segmentation based on Recency, Frequency, and Monetary (RFM) analysis, while implementing the best-performing method in a web-based information system. The dataset consisted of 2,549 valid transactions from 203 unique customers covering October 2024 to October 2025. StandardScaler was applied for data normalization and the elbow method determined the optimal cluster number (k=5). Evaluation using Silhouette Score, Davies-Bouldin Index, and Calinski-Harabasz Index showed K-Means achieved the best results with values of 0.5440, 0.5005, and 282.18, outperforming K-Medoids (0.4790, 0.7034, 145.52) and Hierarchical Clustering (0.5141, 0.5239, 251.73). Five customer segments were identified: Inactive Customers (36.95%), Regular Customers (49.75%), High Value Customers (11.82%), VIP Customers (0.99%), and Top Spenders (0.49%). K-Means was implemented using Streamlit with automatic segmentation and export capabilities to support targeted marketing strategies for each segment.</dc:description>
	<dc:publisher xml:lang="en-US">Informatics Engineering Department of Halu Oleo University</dc:publisher>
	<dc:date>2026-04-26</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://semantik.uho.ac.id/index.php/journal/article/view/275</dc:identifier>
	<dc:identifier>10.55679/semantik.v12i1.275</dc:identifier>
	<dc:source xml:lang="en-US">SemanTIK : Teknik Informasi; Vol. 12 No. 1 (2026): Vol. 12 No. 1 (2026): SemanTIK Teknik Informasi; 22-32</dc:source>
	<dc:source>2502-8928</dc:source>
	<dc:source>2460-1446</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://semantik.uho.ac.id/index.php/journal/article/view/275/60</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2026 Nibertin Zai, Audy Aldrin Kenap, Alfiansyah Hasibuan</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>