Basri, Abdillah Wicaksana (2026) Rancang Bangun Strategi Pemasaran Berbasis Data Dengan Integrasi Efim (Efficient High-Utility Itemset Mining) Pada Odoo Erp: Studi Kasus Supermarket Bahan Bangunan PT XYZ. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6032221005-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6032221005-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy |
Abstract
PT XYZ telah menggunakan Odoo ERP untuk pencatatan transaksi, namun belum untuk analisis strategis, padahal pola pembelian produk bermargin tinggi yang muncul bersamaan pada riwayat data transaksi dapat menjadi sumber rekomendasi cross-selling bagi tim pemasaran. Penelitian ini membangun modul Odoo adv_sale_efim yang menjalankan algoritma Efficient High-Utility Itemset Mining (EFIM) untuk mengidentifikasi kombinasi produk bermargin tinggi dari data transaksi 2024–2025 (239.736 order, 557.266 line). Agar proses analitik tidak mengganggu transaksi operasional, dirancang arsitektur Hybrid Transactional and Analytical Processing (HTAP) yang memisahkan beban OLTP dari beban analitik pada database yang sama. Dengan ambang utility (minutil) 0,15%, EFIM menghasilkan 108 High Utility Itemsets (HUI) sebagai rekomendasi cross-selling yang ditulis langsung ke field native Odoo sehingga muncul otomatis pada Sale Order tanpa penambahan antarmuka tambahan. Pada tiga pair utama, estimasi tambahan margin pada adopsi 25% mencapai Rp 184,3 juta selama dua tahun. Manfaat HTAP terukur pada stabilitas operasional: latensi transaksi Odoo (P95) turun dari 133,2 ms menjadi 34,8 ms saat mining berjalan bersamaan, dengan runtime EFIM 67,75 detik versus 68,56 detik pada skema public. Validitas dibuktikan melalui kesesuaian 100% dengan SPMF. Wawancara dengan dua narasumber internal menyimpulkan rekomendasi EFIM relevan secara bisnis dan arsitektur HTAP layak untuk production. Kontribusi penelitian adalah cetak biru integrasi HUIM pada ERP Odoo untuk strategi pemasaran berbasis margin tanpa mengganggu operasi harian.
========================================================================================================================================
PT XYZ has been using Odoo ERP for recording transactions, yet not for strategic analysis, even though the patterns of high-margin product purchases appearing together in transaction history data can serve as a source of cross-selling recommendations for the marketing team. This research develops the Odoo module adv_sale_efim, which runs the Efficient High-Utility Itemset Mining (EFIM) algorithm to identify combinations of high-margin products from 2024–2025 transaction data (239,736 orders, 557,266 lines). To ensure the analytical process does not disrupt operational transactions, a Hybrid Transactional and Analytical Processing (HTAP) architecture was designed to separate the OLTP workload from the analytical workload on the same database. With a utility threshold (minutil) of 0.15%, EFIM generates 108 High Utility Itemsets (HUI) as cross-selling recommendations written directly to Odoo, making them appear automatically on Sale Orders without requiring any additional interfaces. Across the top three pairs, the estimated additional margin at a 25% adoption rate reaches IDR 184.3 million over two years. The benefits of HTAP are measured in operational stability: Odoo transaction latency (P95) dropped from 133.2 ms to 34.8 ms when mining ran concurrently, with an EFIM runtime of 67.75 seconds versus 68.56 seconds in the public schema. Validity is proven through 100% compliance with SPMF. Interviews with two internal informants concluded that the EFIM recommendations are relevant and the HTAP architecture is viable for production. The contribution of this research is a blueprint for integrating HUIM into Odoo ERP for marginbased marketing strategies without disrupting daily operations.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | EFIM, HUIM, ERP, database, pemasaran berbasis data. |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT) |
| Depositing User: | Abdillah Wicaksana Basri |
| Date Deposited: | 28 Jul 2026 06:46 |
| Last Modified: | 28 Jul 2026 06:46 |
| URI: | http://repository.its.ac.id/id/eprint/139773 |
Actions (login required)
 |
View Item |