CSE Seminar Announcement

Transforming Data into Actionable Knowledge: A data-driven Approach to Enhanced Bridge Deterioration Prediction and Maintenance Decision Making

Wednesday, February 28, 2018
1030 National Center for Supercomputing Applications (NCSA)
12:00pm - 1:00pm
Lunch Provided

Kaijian Liu
CSE Fellow
Civil and Environmental Engineering 

Abstract

Restoring and improving our infrastructure is one of the grand societal challenges; one specific challenge, in the domain of the civil infrastructure system (CIS), is to better predict future bridge deterioration for enhancing our bridge maintenance decision making, so as to optimize maintenance decisions, reduce maintenance cost, and improve bridge safety. The nation’s bridges are in critical conditions: 9% of the bridges are structurally deficient and 14% of them are functionally obsolete. The poor conditions put around 200 million daily trips that are taken across the structurally-deficient bridges into potential safety risks. In order to eliminate the nation’s deficient bridge backlog, we need to invest $20.5 billion in the construction and maintenance of bridges annually; however, only $12.8 billion is being spent currently. Optimal bridge maintenance decisions can only be achieved if we can transform the large amount of increasingly-available bridge data from multiple sources into useful actionable knowledge. There is, thus, an emerging opportunity of leveraging data analytics to better understand the contributing factors to bridge deterioration and the cost-effectiveness of maintenance strategies, and to select and prioritize the operations necessary to maintain the reliability of our bridges.

The goal of this research is to conduct interdisciplinary research towards developing new data-driven, computational tools for extracting and integrating both structured and unstructured bridge data from multiple sources and for analyzing the integrated multi-source bridge data, to transform the large amount of data and information into actionable knowledge for enhancing our abilities to better predict bridge deterioration and to make cost-effective bridge maintenance decisions.