ABSTRACT
Structural Health Monitoring as a nondestructive evaluation tool can be vital for effective bridge maintenance. The advent of wireless sensors and associated technologies has made data collection relatively easy, making continuous monitoring of structures a feasibility. However, it leads to large volumes of data in different formats. This paper presents the data-to-decision framework where data from commercial remote sensing and spatial information technologies are used in a decision-making tool chest (DMT) software with built-in predictive analytical models for making cost-effective predictive maintenance and management decisions. The framework is applied on a selected sub-network of the Union Pacific railroad. The near real-time streams of semi-structured data collected from remote sensors, past visual inspections, and observations from structural analysis are stored in a central data repository. DMT uses this data to assess bridge condition, derive the load capacity, compute reliability indices for bridge components, and provide bridge management recommendations.
