Integrating BMS Economic Evaluation with Random Forest Algorithm for Bridge Maintenance Intelligent Prioritization
DOI:
https://doi.org/10.32832/astonjadro.v15i4.23672Keywords:
BMS, random forest, condition score, economic evaluation, priority ranking.Abstract
Bridges have a long design life, but over time, their function and quality decline due to material degradation and other factors. Therefore, a bridge management system (BMS) is crucial for making appropriate maintenance decisions. This study developed a model to determine bridge maintenance priorities on the Solo-Ngawi Toll Road, which includes 70 bridges by integrating an economic feasibility evaluation (BMS) method with machine learning using the Random Forest algorithm. The bridge condition was conducted in accordance with the 2022 Bridge Inspection Guidelines, while the economic evaluation was based on the 1993 IBMS guidelines, using Net Present Value per square meter (NPV/m²) and Internal Rate of Return (IRR) indicators. The results of the economic evaluation in the form of bridge maintenance priority rankings, were used as the primary target variables for the machine learning model, enabling Random Forest to predict the order of economic priority for maintenance directly. Input variables include 13 bridge element Condition Scores summarized through the Weighted Condition Score (WCS), as well as five economic variables such as remaining service life, floor area, AADT, diversion distance, and maintenance costs. The Random Forest model produces an R² on the test data of 0.8744. Comparison of ranking results between the BMS and Random Forest methods shows good consistency in high-priority bridges and the Random Forest model is able to identify bridges with critical structural conditions that are less accommodated by the conventional BMS method, resulting in more holistic treatment recommendations.
References
[1] F. salam Ahmad, “Dampak Pembangunan Jalan Tol Trans Jawa terhadap Pertumbuhan Ekonomi di Jawa Tengah,” Jurnal Ekonomi dan Kebijakan Pembangunan, vol. 11, no. 1, pp. 1–18, Jul. 2022, doi: 10.29244/jekp.11.1.2022.1-18.
[2] Kementerian PUPR, Pedoman Persyaratan Umum Perencanaan Jembatan. Jakarta, 2015.
[3] Kementerian PUPR, “Panduan dan Rencana Program IBMS,” 1993.
[4] R. A. Dananjoyo, A. Aminullah, and A. S. Budi Nugroho, “Penerapan Metode Life-Cycle Cost dalam Perhitungan Evaluasi Ekonomi Jembatan untuk Penentuan Prioritas Penanganan Jembatan,” Jurnal Teknosains, vol. 9, no. 2, p. 165, Jul. 2020, doi: 10.22146/teknosains.39052.
[5] O. Wahyudhi, A. Aminullah, and A. Triwiyono, “Modifikasi Penilaian pada Sistem Manajemen Jembatan di Indonesia (Interurban Bridge Management System) dengan Mengimplementasikan Condition States dari Metode Bridge Health Index,” 2018.
[6] N. Chawla, “Predicting Bridge Condition Ratings Using Machine Learning and Public Inspection Data,” Nov. 25, 2025. doi: 10.20944/preprints202511.1826.v1.
[7] F. Fard and F. Sadeghi Naieni Fard, “Development and Utilization of Bridge Data of the United States for Predicting Deck Condition Rating Using Random Forest, XGBoost, and Artificial Neural Network,” Remote Sens. (Basel)., vol. 16, no. 2, Jan. 2024, doi: 10.3390/rs16020367.
[8] Y. Wei, R. Ji, Q. Li, and Z. Song, “A Condition Assessment Tool for Steel Bridge Deck Pavement Systems Based on Data Balancing Methods and Machine Learning Algorithms,” Buildings, vol. 14, no. 9, Sep. 2024, doi: 10.3390/buildings14092959.
[9] I. Afriyie, S. N. Joshua, and M. N. Abuanor, “Utilizing Several Machine Learning Techniques to Investigate the Bridge Deck Condition,” Journal of Studies in Civil Engineering, vol. 2, no. 2, pp. 35–53, Dec. 2025, doi: 10.53898/jsce2025223.
[10] B. Wulan Sari and D. Prabowo, “Analisis Perbandingan Prediksi Harga Rumah Dengan Random Forest, Gradient Boosting, dan XGBoost,” Intellect : Indonesian Journal of Learning and Technological Innovation, vol. 4, no. 1, pp. 42–51, Jun. 2025, doi: 10.57255/intellect.v4i1.1385.
[11] M. Lu and S. I. Guler, “Comparison of Random Survival Forest with Accelerated Failure Time-Weibull Model for Bridge Deck Deterioration,” Transp. Res. Rec., vol. 2676, no. 7, pp. 296–311, Jul. 2022, doi: 10.1177/03611981221078281.
[12] F. Hariman, H. H. Christady, A. Triwiyono, B. Kota Sawahlunto, J. Sukarno-Hatta Sawahlunto, and J. Teknik Sipil dan Lingkungan, “Evaluasi dan Program Pemeliharaan Jembatan dengan Metode Bridge Management System (BMS) (Studi Kasus : Empat Jembatan Propinsi D.I. Yogyakarta),” Forum Teknik Sipil No.XVII, 2007.
[13] D. Zindani and S. Singh, “Leveraging Random Forest Regression for Informed and Risk-aware Material Selection in Supercapacitor Fabrication,” Journal of Engineering and Applied Science, vol. 72, no. 1, Dec. 2025, doi: 10.1186/s44147-025-00789-4.
[14] G. E. Pataki and J. H. Boardman, “Bridge Inventory Manual,” New York, 2004.
[15] H. Vaza, R. Sastrawiria, H. Halim, and Septinurriandiani, Identifikasi Kerusakan & Penentuan Nilai Kondisi Jembatan untuk Mendukung Manajemen Aset Jembatan, 1st ed. Jakarta: Pusat Litbang Jalan dan Jembatan Kementerian PUPR, 2017.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 ASTONJADRO

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Paper submitted to ASTONJADRO is the sole property of the Astonjadro Journal. Unless the author withdraws the paper because he does not want to be published in this journal. The publication rights are in the journal Astonjadro.ASTONJADRO
LICENSE
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at http://ejournal.uika-bogor.ac.id/index.php/ASTONJADRO










