Prediksi Harga Pemenang Tender Menggunakan Algoritma Machine Learning (Studi Kasus: PT BKI)

Putri, Vaskya Nabila (2026) Prediksi Harga Pemenang Tender Menggunakan Algoritma Machine Learning (Studi Kasus: PT BKI). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026221128-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Proses penentuan penawaran harga tender di perusahaan penyedia jasa seperti PT Biro Klasifikasi Indonesia (BKI) Cabang Balikpapan saat ini masih menggunakan metode konvensional yang memakan waktu, yang mengakibatkan estimasi harga kurang optimal. Berdasarkan data historis, win rate rata-rata perusahaan hanya mencapai 37,50% dengan deviasi harga sebesar 56,53% dari harga pemenang. Penelitian ini bertujuan untuk mengevaluasi kinerja algoritma Machine Learning dalam memprediksi harga pemenang tender guna mendukung penyusunan harga penawaran yang lebih efisien. Tiga algoritma regresi, yaitu Linear Regression, Support Vector Regression, dan Random Forest Regressor, dibandingkan kinerjanya menggunakan metrik evaluasi Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), dan Mean Absolute Percentage Error (MAPE). Data historis yang digunakan berjumlah 2.622 baris data. Setelah melalui tahap prapemrosesan, diperoleh 673 data yang digunakan sebagai dataset regresi final. Hasil evaluasi menunjukkan fitur harga estimasi (tender_price) menjadi faktor paling dominan dalam memprediksi harga pemenang (winner_price). Model SVR menjadi model machine learning terbaik dengan MAPE 10,83% pada random split dan 8,3% pada temporal split. Hasil tersebut menunjukkan bahwa machine learning dapat digunakan sebagai alat bantu prediksi harga pemenang tender, namun harga estimasi tender tetap menjadi acuan yang sangat kuat dalam memprediksi harga pemenang. Penelitian ini juga menghasilkan prototipe web rule-based parametric calculator untuk membantu perusahaan menyusun biaya penawaran tender dengan lebih terstruktur, objektif, kompetitif, dan berbasis data.
==================================================================================================================================
The process of determining tender bid prices in service provider companies such as PT Biro Klasifikasi Indonesia (BKI) Balikpapan Branch is currently still carried out using conventional methods that are time‑consuming, leading to suboptimal price estimates. Based on historical data, the company’s average win rate is only 37.50%, with a price deviation of 56.53% from the winning price. This study aims to evaluate the performance of machine learning algorithms in predicting the winning tender price to support a more efficient bid price estimation process. Three regression algorithms, namely Linear Regression, Support Vector Regression, and Random Forest Regressor, were compared in terms of their performance using evaluation metrics Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). The historical dataset used in this study consists of 2,622 data rows. After the preprocessing stage, 673 data rows were obtained and used as the final regression dataset. The evaluation results show that the estimated tender price feature (tender_price) is the most dominant factor in predicting the winning price (winner_price). The SVR model was selected as the best machine learning model, achieving a MAPE of 10.83% on the random split and 8.3% on the temporal split. These results indicate that machine learning can be used as a supporting tool for predicting winning tender prices. However, the estimated tender price remains a very strong reference in predicting the winning price. This study also produced a web-based rule-based parametric calculator prototype to help the company structure tender bid cost components in a more systematic, objective, competitive, and data‑driven manner.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Prediksi Harga Tender, Machine Learning, Linear Regression, Random Forest Regressor, Support Vector Regression, Tender Price Prediction
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Vaskya Nabila Putri
Date Deposited:	24 Jul 2026 06:22
Last Modified:	24 Jul 2026 06:22
URI:	http://repository.its.ac.id/id/eprint/137128

Actions (login required)

View Item