The convergence of civil infrastructure and electrical power systems within smart city frameworks necessitates robust, cross-domain monitoring strategies. While machine learning (ML) has shown promise in isolated Structural Health Monitoring (SHM) and Predictive Maintenance (PdM), comparative evaluations across both domains remain fragmented. This study presents a comprehensive comparative analysis of four prominent ML algorithms Random Forest (RF), Support Vector Machines (SVM), Long Short-Term Memory (LSTM) networks, and XGBoost applied to multimodal sensor data. We utilized a synthesized dataset comprising vibration signatures from civil structures (bridge decks) and thermal-electrical load profiles from substation transformers. Our findings indicate that while LSTM networks excel in capturing temporal dependencies in electrical load forecasting (achieving an F1-score of 0.94), tree-based ensemble methods, specifically XGBoost, demonstrate superior efficacy in classifying structural damage from high-dimensional vibration features (accuracy of 96.2%). Furthermore, RF offered the most computationally efficient inference, making it highly suitable for edge-deployment in resource-constrained IoT nodes. This paper provides a practical decision-making framework for civil and electrical engineers selecting ML architectures for integrated smart infrastructure monitoring.
M. el-sseid, L. B. Ben Dalla, Tasnem ELsseid et al.· Al-Farooq Journal of Science...· 0 citations
The Sabha–Ash-Shuwayrif Highway is one of Libya's most strategically important transportation corridors, providing the principal connection between the southern region, the capital Tripoli, and international border crossings. Continuous exposure to heavy freight traffic, extreme thermal fluctuations, and localized moisture intrusion has accelerated asphalt pavement deterioration, resulting in frequent pothole formation, increased maintenance costs, and compromised road safety. Conventional reactive maintenance practices are insufficient for preserving pavement performance because interventions are typically performed only after severe structural damage has occurred. This study proposes an Artificial Intelligence (AI)-based predictive pavement management framework that integrates computer vision, deep learning, and predictive analytics to enable proactive pavement monitoring and maintenance. The proposed framework combines high-resolution pavement imagery, 3D laser profiling, Ground Penetrating Radar (GPR), traffic loading records, Pavement Condition Index (PCI), and environmental data to establish a comprehensive multi-modal dataset. A Convolutional Neural Network (CNN) is employed to automatically detect and classify early-stage pavement distresses, including micro-cracks, alligator cracking, and incipient potholes, while Focal Loss is incorporated to improve the detection of minority distress classes. A Long Short-Term Memory (LSTM) network models the temporal evolution of pavement deterioration using historical PCI, equivalent single axle loads, temperature variations, and precipitation data to forecast future pavement conditions. The predicted deterioration is integrated into an AI-driven risk assessment and decision-support system that prioritizes maintenance activities, recommends appropriate rehabilitation treatments, and optimizes intervention timing according to predicted distress severity. Furthermore, a continuous feedback mechanism updates the predictive models using newly acquired field observations, enabling adaptive learning and long-term performance improvement. The proposed framework is expected to enhance early pothole detection accuracy, reduce lifecycle maintenance costs, improve traffic safety, extend pavement service life, and support data-driven infrastructure management for the Sabha and Ash-Shuwayrif Highway and other strategic transport corridors operating under similar environmental and traffic conditions.
Hana Farhat, Mohammed Hamad, Llahm Omar et al.· Al-Farooq Journal of Science...· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
