This paper provides performance results for a long-term structural health monitoring system developed for a 240 m long reinforced concrete highway bridge using the Internet of Things technology. The design uses multiple sensor technologies (accelerometer, strain and temperature sensors) connected via wireless mesh networks and continuously records the structure’s response over a two-year test period. It can be shown that the proposed solution provides high levels of communication reliability and low transmission latency when assessed in real-world conditions. The use of multi-sensor data fusion techniques effectively detected structural anomalies, and damage identification accuracy was improved with machine learning classification. Controlled simulated damage tests demonstrated that vibration-based indicators are highly responsive to changes in the structure. The use of edge computing reduced both the amount of data that was sent between the monitoring unit and cloud systems and also lowered the overall operational costs associated with monitoring systems. The findings from this study indicate that a low-cost, IoT-based structural health monitoring framework is an effective, practical method for assessing infrastructure condition and managing infrastructure assets over time.