Structural Health Monitoring (SHM) can be a vital tool for effective bridge management. Combining large data sets from multiple sources to create a data-driven decision-making framework is crucial for the success of SHM. This paper presents a big data analytics framework that combines multiple data sets correlated with functional relatedness to convert data into actionable information that empowers risk-based decision-making. The integrated data environment incorporates near real-time streams of semi-structured data from remote sensors, historical visual inspection data, and observations from structural analysis models to monitor, assess, and manage risks associated with the aging bridge inventories. Accelerated processing of dataset is made possible by four technologies: cloud computing, relational database processing, support from NOSQL database, and in-memory analytics. The framework is being validated on a railroad corridor that can be subjected to multiple hazards. The framework enables to compute reliability indices for critical bridge components and individual bridge spans. In addition, framework includes a risk-based decision-making process that enumerate costs and consequences of poor bridge performance at span- and network-levels when rail networks are exposed to natural hazard events such as floods and earthquakes. Big data and high-performance analytics enable insights to assist bridge owners to address problems faster.

Date Published: 20 April 2016
PDF: 7 pages
Proc. SPIE 9803, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2016, 980315 (20 April 2016); doi: 10.1117/12.2219406

Published in SPIE Proceedings Vol. 9803:
Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2016
Jerome P. Lynch, Editor(s)