The primary goal of this proposal is to develop different Machine Learning (ML) algorithms for the rapid identification of damage in bridge structures using the bridge’s dynamic response during regular service operation.  These algorithms are in theory applicable to any dynamical system but will be tailored specifically for bridge structures. This research will provide diagnostic tools that can be directly used in real time by bridge owners for rapid damage assessment.  By collecting data and analyzing them in near real time, the algorithms should be able to provide information on the conditions of the structure and this could help in 1) controlling the traffic operation on the bridge as well as 2) in prioritizing resources in terms of rehabilitation/maintenance operations.

The intended outcome of the project is the development of Machine Learning based tools for rapid damage assessment in bridges which are expected to have tremendous implications in practice and education of modern civil engineers.