Chapter 1: Overview of machine learning in civil engineering

Machine learning (ML) integration with civil engineering has transformed traditional methods by developing predictive models along with automated systems and data-based choice processes in multiple subdisciplines. This research explores all aspects of ML technology in structural health monitoring, geotechnical engineering, transportation systems, water resources management and construction management. ML-driven methods, which include artificial neural networks, support vector machines, decision trees and deep learning models, enhance the processing of structural failures as well as traffic flow optimisation, geotechnical evaluations and hydrological predictions. The progress allows engineers to work with large datasets to discover important information which helps them optimise their infrastructure design and maintenance practices. Modern health monitoring systems based on ML technologies help maintain bridge and building safety through their capabilities to detect material deterioration and forecast structural failures. Geological engineering obtains benefits through automated ML models which perform soil evaluations along with slope assessments and foundation examinations to lower the need for time-consuming testing methods. The application of ML in transportation engineering removes limitations from traffic signal regulation and automatic vehicle navigation while providing timely identification of accidents and enhanced city mobility standards. The practice of water resource engineering applies ML to develop forecasts for floods and conduct hydrological modelling as well as monitor water quality to handle climate-caused risks. Through ML-powered construction management, costs become more predictable, projects are better scheduled and workers become more productive, thus providing better operational efficiency with shortened project times. Civil engineering encounters several barriers when adopting ML technology including a shortage of available data and limitations from explainable models along with hardware capacity and software compatibility issues with ongoing operations. Future development requires dealing with these issues with explainable AI and hybrid ML models and real-time data analytics solutions.