In recent years, the built assets industry has undergone a transformative change with the use of predictive modeling and digital twin technology. The use of data, statistical algorithms and digital twins has become more important to optimize asset management strategies, reduce costs and improve overall asset performance. These advanced techniques can predict asset failure or deterioration, optimize maintenance schedules and minimize downtime, by predicting the future state of an infrastructure and enabling real-time data integration, calculation and decision-making via digital twin platforms. In bridge engineering, the accuracy of deterioration and predictive maintenance models, implemented with digital twin technology, is essential for the efficiency of bridge management systems (BMS), where decisions are made to ensure infrastructure safety while minimizing costs. My research focuses on the study of predictive maintenance models and their integration with digital twin technology in the construction industry, with a particular emphasis on bridge engineering. The research focuses on the lateral bending of the floor beams of an existing aging bridge, where strain and temperature sensors have been installed. Predictive methods, including historical analysis, will be employed on the data collected by these sensors, to forecast structural conditions and the necessary maintenance actions.