An AI-Driven Predicative Frameworks for Sustainable Road Infrastructure and Annual Budget Forecasting in District Level Urban Planning
DOI:
https://doi.org/10.32350.umt-air.61.03Keywords:
Predictive analytics, road infrastructure, budget forecasting, machine learning, sustainability, urban planning, LSTM, XGBoostAbstract
The appropriate annual budget forecasting and sustainable road infrastructure management is still a major challenge facing district level urban planning authorities in the world. Conventional methods are based on historical data analysis and subjective decision-making, which results in reactive approaches to maintenance and inefficient budgets. The given research suggests AI-based predictive model combining machine learning algorithms, namely Random Forest, XGBoost, and Long short-term memory (LSTM) networks, to predict road infrastructure deterioration and plan the budget. The model uses multi-source data that encompasses the quality indicators of roads, traffic flow, environmental indicators and past expenditure data. Pandas, scikit-learn, and TensorFlow were used to preprocess and engineer features using Python and evaluate the model and its performance. Experimental outcomes can prove higher predictive accuracy where XGBoost achieves R2 = 0.89, RMSE = 12.3 and MAE = 9.7 accuracy in budget forecasting and LSTM network is able to capture temporal variations in road degradation with 91 percent accuracy. The framework allows schedule proactively scheduled maintenance, lowers costs of infrastructure by 18-25, and enhances the efficiency of budget utilization. The study is also applicable to sustainable urban development because it gives the decision-makers information-driven long-term infrastructure management, resource optimization and environmental sustainability at the district level.
Downloads
References
[1] World Bank, World Bank Annual Report 2023: A New Era in Development. Washington, DC, USA: World Bank, 2023, doi: https://doi.org/10.1596/AR2023EN.
[2] B. Flyvbjerg, M. K. Skamris Holm, and S. L. Buhl, “How common and how large are cost overruns in transport infrastructure projects?,” Transp. Rev., vol. 23, no. 1, pp. 71–88, 2003, doi: https://doi.org/10.1080/01441640309904.
[3] F. Wang, C.-C. B. Lee, and N. G. Gharaibeh, “Network-level bridge deterioration prediction models that consider the effect of maintenance and rehabilitation,” J. Infrastruct. Syst., vol. 28, no. 1, Art. no. 05021009, Mar. 2022, doi: https://doi.org/10.1061/(ASCE)IS.1943-555X.0000662.
[4] G. Morcous, “Performance prediction of bridge deck systems using Markov chains,” J. Perform. Constr. Facil., vol. 20, no. 2, pp. 146–155, May 2006, doi: https://doi.org/10.1061/(ASCE)0887-3828(2006)20:2(146).
[5] A. Shtayat, S. Moridpour, B. Best, and M. Abuhassan, “Using supervised machine learning algorithms in pavement degradation monitoring,” Int. J. Transp. Sci. Technol., vol. 12, no. 2, pp. 628–639, June 2023, doi: https://doi.org/10.1016/j.ijtst.2022.10.001.
[6] H. Majidifard, Y. Adu-Gyamfi, and W. G. Buttlar, “Deep machine learning approach to develop a new asphalt pavement condition index,” Constr. Build. Mater., vol. 247, Art. no. 118513, June 2020, doi: https://doi.org/10.1016/j.conbuildmat.2020.118513.
[7] A. A. Ali, A. Milad, A. Hussein, N. I. M. Yusoff, and U. Heneash, “Predicting pavement condition index based on the utilization of machine learning techniques: A case study,” J. Road Eng., vol. 3, no. 3, pp. 266–278, Sep. 2023, doi: https://doi.org/10.1016/j.jreng.2023.04.002.
[8] N. Elshaboury, M. S. Yamany, S. Labi, and O. Smadi, “Enhancing local road pavement condition prediction using Bayesian-optimized ensemble machine learning and adaptive synthetic sampling technique,” Int. J. Pavement Eng., vol. 25, no. 1, Art. no. 2365957, 2024, doi: https://doi.org/10.1080/10298436.2024.2365957.
[9] T. Yu, L.-I. Pei, W. Li, Z.-Y. Sun, and J. Huyan, “Pavement surface condition index prediction based on random forest algorithm,” J. Highw. Transp. Res. Dev. (Engl. Ed.), vol. 15, no. 4, pp. 1–11, Dec. 2021, doi: https://doi.org/10.1061/JHTRCQ.0000794.
[10] R. Assaad and I. H. El-Adaway, “Bridge infrastructure asset management system: Comparative computational machine learning approach for evaluating and predicting deck deterioration conditions,” J. Infrastruct. Syst., vol. 26, no. 3, Art. no. 04020032, Sep. 2020, doi: https://doi.org/10.1061/(ASCE)IS.1943-555X.0000572.
[11] Y. Liao, R. Lin, R. Zhang, and G. Wu, “Attention-based LSTM (AttLSTM) neural network for seismic response modeling of bridges,” Comput. Struct., vol. 275, Art. no. 106915, Jan. 2023, doi: https://doi.org/10.1016/j.compstruc.2022.106915.
[12] G. H. Coffie and S. K. F. Cudjoe, “Using extreme gradient boosting (XGBoost) machine learning to predict construction cost overruns,” Int. J. Constr. Manag., vol. 24, no. 16, pp. 1742–1750, 2024, doi: https://doi.org/10.1080/15623599.2023.2289754.
[13] J. Zhang, J. Yuan, A. Mahmoudi, W. Ji, and Q. Fang, “A data-driven framework for conceptual cost estimation of infrastructure projects using XGBoost and Bayesian optimization,” J. Asian Archit. Build. Eng., vol. 24, no. 2, pp. 751–774, 2025, doi: https://doi.org/10.1080/13467581.2023.2294871.
[14] J. van Remmerden, M. Kenter, D. M. Roijers, C. Andriotis, Y. Zhang, and Z. Bukhsh, “Deep multi-objective reinforcement learning for utility-based infrastructural maintenance optimization,” Neural Comput. Appl., vol. 37, pp. 24719–24742, 2025, doi: https://doi.org/10.1007/s00521-024-10954-0.
[15] S. E. Bibri, J. Huang, S. K. Jagatheesaperumal, and J. Krogstie, “The synergistic interplay of artificial intelligence and digital twin in environmentally planning sustainable smart cities: A comprehensive systematic review,” Environ. Sci. Ecotechnol., vol. 20, Art. no. 100433, July 2024, doi: https://doi.org/10.1016/j.ese.2024.100433.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Saba Memon
This work is licensed under a Creative Commons Attribution 4.0 International License.
UMT-AIR follow an open-access publishing policy and full text of all published articles is available free, immediately upon publication of an issue. The journal’s contents are published and distributed under the terms of the Creative Commons Attribution 4.0 International (CC-BY 4.0) license. Thus, the work submitted to the journal implies that it is original, unpublished work of the authors (neither published previously nor accepted/under consideration for publication elsewhere). On acceptance of a manuscript for publication, a corresponding author on the behalf of all co-authors of the manuscript will sign and submit a completed the Copyright and Author Consent Form.
