Damage life cycle analysis for present and future condition assessments using statistical and machine learning techniques

In-service structures experience many changes and damages during construction and operation, affecting their serviceability and remaining life. Infrastructure assessment protocols require regular evaluation of a given structure for a variety of defects and aging phenomena. While there has been extensive research on improving data collection using Non-Destructive Evaluation (NDE) methods , the state of art is limited with regards to NDE data analysis with regards to damage quantification and multi-modal data integration. The purpose of this study is to provide an integrated framework for NDE data assessment including, damage detection and quantification, data correlation, and data fusion. Such analysis initially detects and quantify damages and then the damages are correlated to understand the relation between various measurement techniques. Finally, multi-modal data fusion combines the results of separate NDE methods to improve the assessment of condition ratings. This approach to NDE data analysis provides new and more reliable damage analysis capabilities and a more comprehensive understanding of a damaged structure’s condition, thereby improving decision-making for asset management. The individual aspects of this analytical framework were evaluated through a combination of laboratory and field experiments, yielding promising results.