Article
SMART MONITORING OF URBAN INFRASTRUCTURE USING IOT AND PREDICTIVE ANALYTICS
Recent urban growth and an aging urban infrastructure have made the safe and efficient management of the roads, bridges, buildings, drainage systems and water distribution networks quite difficult. Traditional inspections mainly rely on field visits and manual observation, but this approach is likely to fail to detect early signs of deterioration. The proposed solution is a smart and adaptive sensing system which integrates IoT sensors with predictive analytics for the continuous monitoring of city infrastructure. The envisioned system consists of a cloud-based monitoring platform and penetrators equipped with a network of distributed sensors to measure strain, vibration, crack width, displacement, temperature, moisture, water level, flow, and pressure. An urban infrastructure dataset was developed using simulation to study the proposed approach in both normal, warning and critical operating conditions. Abnormal behaviour was detected from sensor observations and probability of infrastructure deterioration was estimated by using predictive modelling. The analysis demonstrated that a continuous IoT-based monitoring system would be able to detect abnormal structural service conditions before periodic inspections. The combined time-series and condition-based variables was the most successful analytical approach evaluated, with the best classification performance. The study findings suggest that, when combined with the benefits of sensing technology, cloud communication and predictive analytics can help to speed up preventive maintenance, enhance infrastructure safety, minimize over- or underinspections and allow civil engineering authorities to more effectively apply their maintenance resources.
Full Text Attachment
