Lisken Ratna Kinasih, . (2026) SEGMENTASI PRODUK MULTI TOKO BERDASARKAN PENJUALAN DAN PROFITABILITAS DENGAN ALGORITMA K-MEANS SEBAGAI PENDUKUNG KEPUTUSAN BISNIS PADA PT XYZ. Tugas Akhir thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (83kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (191kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (831kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (280kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (5MB) |
|
Text
BAB 5.pdf Download (167kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (167kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository staff only Download (336kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (370kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (15MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (2MB) |
Abstract
The growth of e-commerce drives businesses to make decisions based on sales data. With varying product success in terms of sales and profitability, PT XYZ operates three online stores: Yoiki, Xinbie, and Kanata. The K-Means Clustering technique will be used in this project to segment multi-store items, and the findings will be incorporated into an interactive web dashboard. Sales data from October 2025 to March 2026 and Cost of Goods Sold (HPP) data were utilized. The Knowledge Discovery in Database (KDD) step of the data processing process is followed by clustering attributes such as quantity sold, total revenue, total profit, and margin (%). The Davies-Bouldin Index (DBI) is used to assess the clustering findings, the Elbow approach is used to identify the ideal number of clusters, and MinMaxScaler is used to normalize the data. The findings indicate that K = 3 for all shops, K = 3 for Yoiki, K = 3 for Xinbie, and K = 2 for Kanata are the ideal number of clusters. Products are divided into three categories based on the segmentation results: superior products, medium or prospective products, and less potential items. Businesses may utilize the segmentation information to assess underperforming items, improve marketing methods, and prioritize inventory and promotions. Built using Streamlit and SQLite, the website system can manage HPP data, display dashboards, perform clustering, save clustering history, and present segmentation results. Black box testing revealed that the system's primary functions operated as anticipated.
| Item Type: | Thesis (Tugas Akhir) |
|---|---|
| Additional Information: | [No.Panggil: 2310501080] [Pembimbing: Ika Nurlaili Isnainiyah] [Penguji 1: Nur Hafifah Matondang] [Penguji 2: M. Bayu Wibisono] |
| Uncontrolled Keywords: | Product Segmentation, K-Means, Clustering, Profitability, Streamlit |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Sistem Informasi (DIII) |
| Depositing User: | LISKEN RATNA KINASIH |
| Date Deposited: | 24 Aug 2026 03:44 |
| Last Modified: | 24 Aug 2026 03:44 |
| URI: | http://repository.upnvj.ac.id/id/eprint/51265 |
Actions (login required)
 |
View Item |