Implementação de Modelos de Predição de Riscos Baseados em Aprendizado de Máquina para Projetos de Construção de Infraestruturas em Larga Escala em Ambientes Urbanos

Autores

  • Ali Hussain Alhamami Civil Engineering Department, College of Engineering. Najran University, Najran, Kingdom of Saudi Arabia ahalhamami@nu.edu.sa https://orcid.org/0009-0007-1455-2315

DOI:

https://doi.org/10.14571/brajets.v18.n1.330-346

Palavras-chave:

predição de riscos na construção, árvores de decisão com reforço de gradiente, projetos de infraestrutura, aplicações de aprendizado de máquina, gestão da construção urbana

Resumo

Projetos de infraestrutura em larga escala geralmente utilizam abordagens reativas para gerenciar riscos na construção, o que pode levar a atrasos custosos e aumentos de orçamento. Este estudo cria e testa uma estrutura de predição de riscos que utiliza aprendizado de máquina, especificamente árvores de decisão com reforço de gradiente (GBDT), para ajudar a identificar e lidar com riscos nas fases iniciais da construção de infraestrutura urbana. Foram analisados dados de 220 projetos de infraestrutura, abrangendo o período de 2015 a 2024 e localizados na América do Norte, Europa e Ásia. Esses projetos tinham valores entre 50 milhões e 2 bilhões de dólares. A abordagem combinou análise de componentes principais e GBDT, lidando com 47 variáveis relacionadas a riscos em seis áreas distintas. Para testar o modelo, foi utilizada validação cruzada em 5 etapas, juntamente com validação temporal, que consistiu em deixar de fora os 20% mais recentes dos projetos. O modelo GBDT atingiu uma precisão geral de predição de 87,3%, superando os métodos tradicionais em 23%. A capacidade de detectar riscos precocemente melhorou significativamente, de 45% para 78%, resultando em uma redução média de custos de 12,4%. Os riscos técnicos apresentaram a maior precisão de predição, com 89,4%, enquanto a otimização de recursos teve uma melhoria de 25,7% no uso dos equipamentos. Considera-se que essa estrutura baseada em aprendizado de máquina melhora significativamente a gestão de riscos na construção, oferecendo maior precisão, detecção precoce de riscos e economia de custos, sugerindo seu uso amplo na construção de infraestrutura urbana.

Referências

Aljohani A., (2023), Predictive analytics and machine learning for real-time supply chain risk mitigation and agility, Sustainability, 15, 15088. DOI: https://doi.org/10.3390/su152015088

Alvand A., Mirhosseini S.M., Ehsanifar M., Zeighami E., Mohammadi A., (2023), Identification and assessment of risk in construction projects using the integrated FMEA-SWARA-WASPAS model under fuzzy environment: a case study of a construction project in Iran, International Journal of Construction Management, 23, 392–404. https://doi.org/10.1080/15623599.2021.1877875 DOI: https://doi.org/10.1080/15623599.2021.1877875

Ashtari M.A., Ansari R., Hassannayebi E., Jeong J., (2022), Cost overrun risk assessment and prediction in construction projects: A Bayesian network classifier approach, Buildings, 12, 1660. DOI: https://doi.org/10.3390/buildings12101660

Ayubi Rad M., Ayubirad M.S., (2017), Comparison of artificial neural network and coupled simulated annealing based least square support vector regression models for prediction of compressive strength of high-performance concrete, Scientia Iranica, 24, 487–496. DOI: https://doi.org/10.24200/sci.2017.2412

Bahamid R.A., Doh S.I., Khoiry M.A., Kassem M.A., Al-Sharafi M.A., (2022), The current risk management practices and knowledge in the construction industry, Buildings, 12, 1016. DOI: https://doi.org/10.3390/buildings12071016

Cardellicchio A., Ruggieri S., Nettis A., Renò V., Uva G., (2023), Physical interpretation of machine learning-based recognition of defects for the risk management of existing bridge heritage, Engineering Failure Analysis, 149, 107237. DOI: https://doi.org/10.1016/j.engfailanal.2023.107237

Chenya L., Aminudin E., Mohd S., Yap L.S., (2022), Intelligent risk management in construction projects: Systematic literature review, Ieee Access, 10, 72936–72954. DOI: https://doi.org/10.1109/ACCESS.2022.3189157

Chew A.W.Z., He R., Zhang L., (2025), Physics Informed Machine Learning (PIML) for Design, Management and Resilience-Development of Urban Infrastructures: A Review, Archives of Computational Methods in Engineering, 32, 399–439. https://doi.org/10.1007/s11831-024-10145-z DOI: https://doi.org/10.1007/s11831-024-10145-z

Dar I.S., Chand S., Shabbir M., Kibria B.G., (2023), Condition-index based new ridge regression estimator for linear regression model with multicollinearity, Kuwait Journal of Science, 50, 91–96. DOI: https://doi.org/10.1016/j.kjs.2023.02.013

De Santis E., Arnò F., Rizzi A., (2022), Estimation of fault probability in medium voltage feeders through calibration techniques in classification models, Soft Computing, 26, 7175–7193. https://doi.org/10.1007/s00500-022-07194-6 DOI: https://doi.org/10.1007/s00500-022-07194-6

Di Sante M., Mazzieri F., Fratalocchi E., (2021), RECENT DEVELOPMENTS IN SITE SPECIFIC RISK ASSESSMENT FOR POLLUTED SITES, P-ESEM, 629.

Freddi F., Galasso C., Cremen G., Dall’Asta A., Di Sarno L., Giaralis A., Gutiérrez-Urzúa F., Málaga-Chuquitaype C., Mitoulis S.A., Petrone C., (2021), Innovations in earthquake risk reduction for resilience: Recent advances and challenges, International Journal of Disaster Risk Reduction, 60, 102267. DOI: https://doi.org/10.1016/j.ijdrr.2021.102267

Garcia J., Villavicencio G., Altimiras F., Crawford B., Soto R., Minatogawa V., Franco M., Martínez-Muñoz D., Yepes V., (2022), Machine learning techniques applied to construction: A hybrid bibliometric analysis of advances and future directions, Automation in Construction, 142, 104532. DOI: https://doi.org/10.1016/j.autcon.2022.104532

Ghasemi, S., Meybodi, M. R., Fooladi, M. D. T., & Rahmani, A. M. (2018). A cost-aware mechanism for optimized resource provisioning in cloud computing. Cluster Computing, 21, 1381-1394. DOI: https://doi.org/10.1007/s10586-017-1271-z

Gondia A., Ezzeldin M., El-Dakhakhni W., (2022), Machine Learning–Based Decision Support Framework for Construction Injury Severity Prediction and Risk Mitigation, ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 8, 04022024. https://doi.org/10.1061/AJRUA6.0001239 DOI: https://doi.org/10.1061/AJRUA6.0001239

Hakiri A., Gokhale A., Yahia S.B., Mellouli N., (2024), A comprehensive survey on digital twin for future networks and emerging Internet of Things industry, Computer Networks, 110350. DOI: https://doi.org/10.1016/j.comnet.2024.110350

Jiang H., Guo H., Sun Z., Xing Q., Zhang H., Ma Y., Li S., (2022), Projections of urban built-up area expansion and urbanization sustainability in China’s cities through 2030, Journal of Cleaner Production, 367, 133086. DOI: https://doi.org/10.1016/j.jclepro.2022.133086

Kaur, K., Kim, D., Jamshidi, A., & Zhang, L. (2025). Identifying Flaky Tests in Quantum Code: A Machine Learning Approach. arXiv preprint arXiv:2502.04471.

Khodabakhshian A., Puolitaival T., Kestle L., (2023), Deterministic and probabilistic risk management approaches in construction projects: A systematic literature review and comparative analysis, Buildings, 13, 1312. DOI: https://doi.org/10.3390/buildings13051312

Koh K.-Y., Ahmad S., Lee J., Suh G.-H., Lee C.-M., (2022), Hierarchical clustering on principal components analysis to detect clusters of highly pathogenic avian influenza subtype H5N6 epidemic across South Korean Poultry Farms, Symmetry, 14, 598. DOI: https://doi.org/10.3390/sym14030598

Leanza A., Bonanno S., Suriano E., Amara G., Gigli C., (2017), The SWOT analysis applied to a high risk area as a strategy to increase sustainable local value chain, Procedia Environmental Science, Engineering and Management, 4, 69–76.

Li Q., Chen Q., Wu J., Qiu Y., Zhang C., Huang Y., Guo J., Yang B., (2023), XGBoost-based intelligent decision making of HVDC system with knowledge graph, Energies, 16, 2405. DOI: https://doi.org/10.3390/en16052405

Love P.E., Ika L.A., Pinto J.K., (2022), Homo heuristicus: From risk management to managing uncertainty in large-scale infrastructure projects, IEEE Transactions on Engineering Management, 71, 1940–1949. DOI: https://doi.org/10.1109/TEM.2022.3170474

Mashali A., Elbeltagi E., Motawa I., Elshikh M., (2023), Stakeholder management challenges in mega construction projects: critical success factors, Journal of Engineering, Design and Technology, 21, 358–375. DOI: https://doi.org/10.1108/JEDT-09-2021-0483

Mazher K.M., Chan A.P., Choudhry R.M., Zahoor H., Edwards D.J., Ghaithan A.M., Mohammed A., Aziz M., (2022), Identifying measures of effective risk management for public–private partnership infrastructure projects in developing countries, Sustainability, 14, 14149. DOI: https://doi.org/10.3390/su142114149

McDermot E., Agdas D., Rodríguez Díaz C.R., Rose T., Forcael E., (2022), Improving performance of infrastructure projects in developing countries: an Ecuadorian case study, International Journal of Construction Management, 22, 2469–2483. https://doi.org/10.1080/15623599.2020.1797985 DOI: https://doi.org/10.1080/15623599.2020.1797985

O’Sullivan J.N., (2023), Demographic delusions: World population growth is exceeding most projections and jeopardising scenarios for sustainable futures, World, 4, 545–568. DOI: https://doi.org/10.3390/world4030034

Ozaki Y., Tanigaki Y., Watanabe S., Nomura M., Onishi M., (2022), Multiobjective tree-structured parzen estimator, Journal of Artificial Intelligence Research, 73, 1209–1250. DOI: https://doi.org/10.1613/jair.1.13188

Pan Y., Zhang L., (2023), Integrating BIM and AI for Smart Construction Management: Current Status and Future Directions, Archives of Computational Methods in Engineering, 30, 1081–1110. https://doi.org/10.1007/s11831-022-09830-8 DOI: https://doi.org/10.1007/s11831-022-09830-8

Pirgazi J., Kallehbasti M.M.P., Sorkhi A.G., Kermani A., (2024), An efficient hybrid filter-wrapper method based on improved Harris Hawks optimization for feature selection, BioImpacts, 15, 30340–30340. DOI: https://doi.org/10.34172/bi.30340

Pomaza-Ponomarenko A., Kryvova S., Hordieiev A., Hanzyuk A., Halunko O., (2023), Innovative risk management: identification, assessment and management of risks in the context of innovative project management. https://www.indianjournals.com/ijor.aspx?target=ijor:eaj&volume=68&issue=4&article=034 DOI: https://doi.org/10.46852/0424-2513.4.2023.34

Rezvani S.M., Falcão M.J., Komljenovic D., de Almeida N.M., (2023), A systematic literature review on urban resilience enabled with asset and disaster risk management approaches and GIS-based decision support tools, Applied Sciences, 13, 2223. DOI: https://doi.org/10.3390/app13042223

Rising J., Tedesco M., Piontek F., Stainforth D.A., (2022), The missing risks of climate change, Nature, 610, 643–651. DOI: https://doi.org/10.1038/s41586-022-05243-6

Sanni-Anibire M.O., Zin R.M., Olatunji S.O., (2022), Machine learning model for delay risk assessment in tall building projects, International Journal of Construction Management, 22, 2134–2143. https://doi.org/10.1080/15623599.2020.1768326 DOI: https://doi.org/10.1080/15623599.2020.1768326

Sharopova, N. (2023). Enhancing Digital Market Research Through Distributed Data and Knowledge-Based Systems: Analyzing Emerging Trends and Strategies. In International Conference on Next Generation Wired/Wireless Networking (pp. 251-259). Cham: Springer Nature Switzerland. DOI: https://doi.org/10.1007/978-3-031-60997-8_23

Shoar S., Chileshe N., Edwards J.D., (2022), Machine learning-aided engineering services’ cost overruns prediction in high-rise residential building projects: Application of random forest regression, Journal of Building Engineering, 50, 104102. DOI: https://doi.org/10.1016/j.jobe.2022.104102

Siahkouhi M., Rashidi M., Mashiri F., Aslani F., Ayubirad M.S., (2024), Application of self-sensing concrete sensors for bridge monitoring- A review of recent developments, challenges, and future prospects, Measurement, 116543. DOI: https://doi.org/10.1016/j.measurement.2024.116543

Wong L.-W., Tan G.W.-H., Ooi K.-B., Lin B., Dwivedi Y.K., (2024), Artificial intelligence-driven risk management for enhancing supply chain agility: A deep-learning-based dual-stage PLS-SEM-ANN analysis, International Journal of Production Research, 62, 5535–5555. https://doi.org/10.1080/00207543.2022.2063089 DOI: https://doi.org/10.1080/00207543.2022.2063089

Yazdi M., Adumene S., Tamunodukobipi D., Mamudu A., Goleiji E., (2025), Virtual Safety Engineer: From Hazard Identification to Risk Control in the Age of AI, In M. Yazdi (Ed.), Safety-Centric Operations Research: Innovations and Integrative Approaches (Vol. 232, pp. 91–110), Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-82934-5_5 DOI: https://doi.org/10.1007/978-3-031-82934-5_5

Publicado

28-03-2025

Como Citar

Alhamami , A. H. (2025). Implementação de Modelos de Predição de Riscos Baseados em Aprendizado de Máquina para Projetos de Construção de Infraestruturas em Larga Escala em Ambientes Urbanos. Cadernos De Educação, Tecnologia E Sociedade, 18(1), 330–346. https://doi.org/10.14571/brajets.v18.n1.330-346

Edição

Seção

Artigo