Model Prediksi Pembelian Ulang Untuk Strategi Retensi Pelanggan E-Commerce Menggunakan Lrfm Dan Random Forest: Studi Kasus Malda Kids

Wijaya, Wikudhara Shofrina (2026) Model Prediksi Pembelian Ulang Untuk Strategi Retensi Pelanggan E-Commerce Menggunakan Lrfm Dan Random Forest: Studi Kasus Malda Kids. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6032231168-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Usaha Mikro, Kecil, dan Menengah (UMKM) berperan penting dalam perekonomian Indonesia, termasuk melalui pemanfaatan platform e-commerce. Namun, banyak UMKM masih menghadapi tantangan dalam menjaga loyalitas pelanggan karena belum mengoptimalkan pemanfaatan data transaksi. Penelitian ini bertujuan untuk menganalisis perilaku pelanggan dan membangun model prediksi repeat purchase dalam periode 90 hari pada UMKM e-commerce Malda Kids yang bergerak di bidang penjualan busana muslim anak. Pendekatan LRFM (Length, Recency, Frequency, Monetary) digunakan untuk mengekstraksi karakteristik perilaku pelanggan berdasarkan data transaksi historis periode 2022– 2024. Selanjutnya, dilakukan segmentasi pelanggan menggunakan algoritma KMeans dan pemodelan prediksi pembelian ulang menggunakan Random Forest dengan data yang telah melalui proses penyeimbangan kelas. Hasil penelitian menunjukkan bahwa variabel Frequency dan Monetary merupakan faktor paling dominan dalam memengaruhi kemungkinan repeat purchase. Penelitian ini diharapkan dapat mendukung pengambilan keputusan pemasaran dan strategi retensi pelanggan UMKM secara lebih efektif dan terukur.
================================================================================================================================
Micro, Small, and Medium Enterprises (MSMEs) play a crucial role in the Indonesian economy, including through the utilization of e-commerce platforms. However, many MSMEs still face challenges in maintaining customer loyalty due to the underutilization of transaction data. This study aims to analyze customer behavior and develop a predictive model of repeat purchase within a 90-day period for Malda Kids, an MSME e-commerce business specializing in Muslim children’s apparel. The LRFM (Length, Recency, Frequency, Monetary) approach is applied to extract customer behavioral characteristics based on historical transaction data from 2022 to 2024. Customer segmentation is then performed using the K-Means clustering algorithm, followed by repeat purchase prediction using a Random Forest model with balanced class data. The results indicate that Frequency and Monetary are the most dominant variables influencing the likelihood of repeat purchase. This study is expected to support more effective and data-driven marketing decision-making and customer retention strategies for MSME e-commerce businesses.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	LRFM, Repeat Purchase, Random Forest, Segementasi Pelanggan, UMKM e-commerce. LRFM, Repeat Purchase, Random Forest, Customer Segmentation, E-commerce MSMEs.
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT)
Depositing User:	Wikudhara Shofrina Wijaya
Date Deposited:	06 Feb 2026 07:37
Last Modified:	06 Feb 2026 07:37
URI:	http://repository.its.ac.id/id/eprint/132234

Actions (login required)

View Item