OPTIMALISASI ALOKASI DANA HIBAH SEBAGAI MODEL MANAJEMEN KEUANGAN PROYEK

Satriawaty Mallu; Husmaruddin Husmaruddin; Goso Goso

doi:10.59003/nhj.v6i1.2290

Authors

Satriawaty Mallu Universitas Muhammadiyah Palopo
Husmaruddin Husmaruddin Universitas Muhammadiyah Palopo
Goso Goso Universitas Muhammadiyah Palopo

DOI:

https://doi.org/10.59003/nhj.v6i1.2290

Keywords:

grant allocation, project financial management, machine learning, XGBoost, budget optimisation

Abstract

This study aims to develop a grant fund allocation optimization model as part of project financial management using the Extreme Gradient Boosting (XGBoost) algorithm. The research is motivated by the persistent gap between approved grant funds and actual budget realization during project implementation, which potentially leads to either resource wastage or funding shortages. The research methodology involves collecting historical grant fund data, performing data cleaning through missing value handling, outlier detection using the Interquartile Range (IQR) method, and logical consistency validation across variables. Feature engineering was subsequently conducted to generate relevant derived variables, such as budget realization percentage and budget efficiency, prior to model training. The XGBoost model's performance was compared against Linear Regression and Random Forest Regressor using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R² Score as evaluation metrics. The results indicate that the tuned XGBoost model achieved the best performance with an R² Score of 0.912, outperforming Random Forest (0.861) and Linear Regression (0.742). Feature importance analysis revealed that historical performance scores and budget realization percentages were the most dominant factors in predicting optimal grant fund allocation. Model simulation demonstrated the system's capability to provide more objective fund allocation recommendations based on grantees' performance track records. This study concludes that the XGBoost algorithm is effective as a decision support model for grant fund allocation optimization and can be implemented by grant management institutions to enhance the efficiency and accountability of project financial management.

Downloads

Download data is not yet available.

References

Adnan, M., & Rahman, A. (2022). Machine learning approaches for public fund allocation optimization: A systematic review. Journal of Public Administration and Governance, 12(3), 145-162.

Ahmed, S., Khan, R., & Ali, M. (2021). Predictive analytics for grant fund disbursement using ensemble learning methods. International Journal of Financial Studies, 9(4), 78-95.

Anwar, H., & Susanto, B. (2023). Penerapan algoritma XGBoost dalam prediksi risiko kredit pada lembaga keuangan mikro. Jurnal Teknologi Informasi dan Ilmu Komputer, 10(2), 233-244.

Chen, L., & Wang, X. (2021). Comparative analysis of gradient boosting algorithms for financial risk prediction. Applied Sciences, 11(15), 6890-6905.

Chen, Y., Zhao, J., & Liu, H. (2022). A hybrid XGBoost-based model for budget allocation efficiency in public sector projects. Sustainability, 14(8), 4521-4538.

Fadillah, R., & Kurniawan, D. (2023). Optimalisasi alokasi anggaran proyek pemerintah menggunakan pendekatan machine learning. Jurnal Sistem Informasi Bisnis, 13(1), 55-68.

Gunawan, T., & Pratiwi, N. (2022). Model prediksi efisiensi penggunaan dana hibah penelitian berbasis random forest dan XGBoost. Jurnal Ilmu Komputer dan Informasi, 15(2), 112-125.

Hasan, M. K., & Sultana, N. (2023). Application of ensemble machine learning models in nonprofit grant fund management. Journal of Nonprofit and Public Sector Marketing, 35(2), 210-229.

Hidayat, F., & Nugroho, A. (2021). Analisis kinerja algoritma machine learning dalam prediksi realisasi anggaran proyek konstruksi. Jurnal Teknik Sipil dan Perencanaan, 23(1), 45-58.

Jabeur, S. B., Mefteh-Wali, S., & Viviani, J. L. (2021). Forecasting gold price with the XGBoost algorithm and SHAP interpretability analysis. Annals of Operations Research, 1-21.

Kartika, D., & Setiawan, R. (2022). Penerapan data mining untuk optimalisasi alokasi dana bantuan sosial pemerintah daerah. Jurnal Ekonomi dan Kebijakan Publik, 13(2), 189-204.

Kim, J., Park, S., & Lee, K. (2021). Explainable AI for financial resource allocation: An XGBoost and SHAP approach. Expert Systems with Applications, 183, 115-134.

Kusuma, A. P., & Wardani, R. (2023). Prediksi keberhasilan proyek penelitian berbasis hibah menggunakan pendekatan machine learning. Jurnal Manajemen Teknologi, 22(1), 78-92.

Li, X., Zhang, Y., & Sun, W. (2022). Machine learning-based prediction of project cost overruns: Evidence from public infrastructure projects. Journal of Construction Engineering and Management, 148(6), 04022032.

Liu, Y., Chen, Z., & Huang, Q. (2023). A comparative study of tree-based ensemble models for budget forecasting in nonprofit organizations. Journal of Financial Data Science, 5(1), 44-60.

Mahmud, R., & Islam, T. (2021). XGBoost-based prediction model for microfinance loan allocation efficiency. International Journal of Emerging Markets, 16(7), 1345-1362.

Nasution, S., & Aulia, R. (2022). Model manajemen keuangan proyek berbasis data-driven decision making menggunakan machine learning. Jurnal Akuntansi dan Keuangan Indonesia, 19(2), 167-183.

Prasetyo, B., & Wibowo, H. (2023). Implementasi algoritma XGBoost untuk prediksi anggaran optimal pada proyek penelitian dosen. Jurnal Sistem Informasi dan Teknologi Informasi, 12(3), 201-215.

Putra, I. K. G. D., & Sudarma, M. (2022). Implementasi algoritma XGBoost untuk prediksi kinerja keuangan pada perusahaan berbasis data historis. Jurnal Ilmu Komputer dan Informasi, 15(1), 45-53.

Rahmawati, D., & Firmansyah, A. (2021). Analisis efisiensi pengelolaan dana hibah penelitian menggunakan pendekatan data envelopment analysis dan machine learning. Jurnal Riset Akuntansi dan Keuangan, 9(3), 301-315.

Santoso, E., & Ramadhan, F. (2022). Prediksi risiko gagal bayar dana bantuan pemerintah menggunakan algoritma XGBoost. Jurnal Sistem Informasi, 18(1), 33-47.

Setiawan, B., & Yuliana, I. (2023). Optimalisasi anggaran proyek berbasis kecerdasan buatan: Studi kasus lembaga penyalur hibah penelitian. Jurnal Manajemen dan Bisnis, 21(2), 145-160.

Sharma, P., & Gupta, R. (2021). Application of gradient boosting algorithms in public sector financial resource optimization. International Journal of Public Sector Management, 34(5), 512-530.

Sudirman, A., & Lestari, W. (2022). Penerapan machine learning untuk mendukung pengambilan keputusan alokasi dana hibah riset perguruan tinggi. Jurnal Pendidikan dan Teknologi Informasi, 14(2), 98-112.

Wang, Y., Wang, D., Geng, N., Wang, Y., Yin, Y., & Jin, Y. (2020). Stacking-based ensemble learning of decision trees for interpretable prediction of air quality. Applied Soft Computing, 86, 105848.

Wibowo, S., & Handayani, T. (2023). Sistem pendukung keputusan berbasis XGBoost untuk evaluasi proposal hibah penelitian. Jurnal Teknik Informatika dan Sistem Informasi, 9(1), 67-80.

Wu, Z., Lin, W., & Zhang, Z. (2022). An XGBoost-based approach for financial risk assessment and resource allocation optimization in project management. Expert Systems with Applications, 202, 117227.

Yulianto, D., & Prabowo, A. (2021). Analisis komparasi algoritma machine learning dalam prediksi realisasi anggaran hibah penelitian. Jurnal Sains dan Teknologi Informasi, 7(2), 88-101.

Zhang, Y., Zhao, Z., & Zheng, J. (2023). Application of XGBoost model in budget allocation prediction for public research grants. Journal of Financial Data Science, 5(2), 88-102.

Zulkarnain, R., & Anggraini, P. (2022). Model prediktif berbasis ensemble learning untuk efisiensi alokasi dana hibah pendidikan. Jurnal Teknologi dan Manajemen Informatika, 11(2), 122-136.