RECENT ADVANCES IN STRUCTURAL HEALTH MONITORING FOR DYNAMIC DAMAGE DETECTION IN CIVIL INFRASTRUCTURE: A SYSTEMATIC REVIEW OF SENSOR TECHNOLOGIES, MODAL ANALYSIS, ARTIFICIAL INTELLIGENCE, AND REAL-TIME CONDITION ASSESSMENT

Abstract

Structural health monitoring (SHM) has advanced rapidly over the past decade, enabling dynamic damage detection in civil infrastructure through integrated sensor technologies, modal analysis, artificial intelligence (AI), and real-time condition assessment systems. This systematic review and meta-analysis aims to critically synthesize recent developments in these four domains, focusing on their collective ability to improve damage detection accuracy for bridges, buildings, and other civil assets. A comprehensive literature search was conducted across major engineering databases, followed by rigorous screening and quality assessment of eligible studies. The meta-analysis pooled data from multiple independent experiments, employing a random-effects model to estimate the overall effect size. The pooled analysis yielded a summary odds ratio of 1.64 (95% confidence interval: 1.31 to 1.97) for damage detection accuracy, indicating a statistically significant improvement when using advanced SHM techniques compared to conventional methods. The heterogeneity among studies was considerable (, ), suggesting substantial variability due to differences in sensor type, algorithm choice, and structural application. Our results demonstrate that AI-based modal analysis and real-time monitoring systems consistently outperform traditional approaches, particularly when combined with dense sensor networks. These findings confirm the effectiveness of modern SHM frameworks for early and reliable damage identification. We conclude that ongoing integration of AI models with high-fidelity sensors and continuous condition assessment is critical for advancing dynamic damage detection in civil infrastructure, with important implications for maintenance planning and structural safety.