Ictiasha, Firchi (2026) Pengembangan Framework Pendukung Keputusan Jumlah Penyedia Dan Alokasi Order Menggunakan Model Machine Learning Di PT PLN (Persero): Studi Kasus Perencanaan Pengadaan Kontrak Kesepakatan Harga Satuan Transformator Tenaga. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6010241058-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6010241058-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (12MB) | Request a copy |
Abstract
Pengadaan transformator tenaga di PT PLN (Persero) saat ini masih mengandalkan pendekatan Harga Evaluasi Akhir sehingga data kinerja penyedia multidimensi belum dimanfaatkan secara optimal dalam keputusan alokasi order. Penelitian ini mengembangkan kerangka kerja Supplier selection and Order allocation (SSOA) berbasis machine learning menggunakan algoritma Decision Tree, Random Forest, dan XGBoost dengan data historis pengadaan transformator tenaga periode 2023 sampai 2025. Kinerja penyedia dimodelkan berdasarkan dimensi biaya, ketepatan waktu, kualitas, dan risiko kontraktual untuk menghasilkan klasifikasi kinerja penyedia. Hasil evaluasi menunjukkan bahwa Random Forest dipilih sebagai model terbaik dengan nilai macro AUC sebesar 0,9882. Hasil klasifikasi menunjukkan dua penyedia berada pada Tier 1, satu penyedia pada Tier 2, dan tiga penyedia pada Tier 3, serta menghasilkan rekomendasi satu hingga tiga penyedia yang layak untuk setiap varian transformator tenaga. Rekomendasi skenario alokasi konservatif memprioritaskan penyedia berkinerja tinggi dengan alokasi 70% pada Tier 1 dan 30% pada Tier 2. Pendekatan data-driven ini memberikan solusi yang objektif, transparan, dan akuntabel.
=====================================================================================================================================
The procurement of power transformers at PT PLN (Persero) currently still relies on the Final Evaluation Price approach, so multidimensional supplier performance data has not been optimally utilized in order allocation decisions. This research develops a Supplier selection and Order allocation (SSOA) framework based on machine learning using Decision Tree, Random Forest, and XGBoost algorithms with historical data on power transformer procurement for the period 2023 to 2025. Supplier performance is modeled based on the dimensions of cost, timeliness, quality, and contractual risk to produce a classification of supplier performance. The evaluation results show that Random Forest was chosen as the best model with a macro-AUC value of 0.9882. The classification results show that two suppliers are in Tier 1, one supplier in Tier 2, and three suppliers in Tier 3, and generate recommendations of one to three eligible suppliers for each power transformer variant. The conservative allocation scenario recommendation prioritizes high-performing suppliers with a 70% allocation to Tier 1 and 30% to Tier 2. This data-driven approach provides solutions that are objective, transparent, and accountable.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Pengadaan, Supplier selection and Order allocation (SSOA), Machine learning, Kontrak Harga Satuan Transformator Tenaga, PT PLN (Persero).Procurement, Supplier selection and Order allocation (SSOA), Machine learning, Unit Price Contracts for Power Transformers, PT PLN (Persero). |
| Subjects: | T Technology > T Technology (General) > T58.62 Decision support systems |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26101-(S2) Master Thesis |
| Depositing User: | Firchi Ictiasha |
| Date Deposited: | 27 Jul 2026 04:50 |
| Last Modified: | 27 Jul 2026 04:50 |
| URI: | http://repository.its.ac.id/id/eprint/137892 |
Actions (login required)
 |
View Item |