Evaluation of non-destructive impact-echo data from the national bridge inventory

Abstract

The National Bridge Inventory's free and public database InfoBridge contains a large quantity of non-destructive test (NDT) data. We studied the InfoBridge database, focusing specifically on 590 000 impact-echo test data that are housed there, to identify challenges that need to be addressed in order to enable effective big data analysis. The impact-echo data studied were collected by three different test devices – RABIT, Cane, and Stepper – from 78 different surveys at 38 bridge decks between 2009 and 2019. Analysis of all impact-echo data and reports shows that the three test devices provide different data quality, with Cane having the best overall data quality. The data quality of RABIT was poor in 2017, while it was better in the years before and after. Impact-echo maps collected from successive measurement campaigns show that condition indices derived from the impact-echo maps are not reproducible between test devices and surveys.

Keywords

• Concrete testing
• Bridge deck
• Defect detection
• Impact-echo
• Nondestructive testing
• Vibration