Data Driven Predictive Analytics for Bridge Asset Management is normally used for suitable repair and or rehabilitation plan of a typical bridge system based on the development of degradation model. Bridges would be showing signs of distress due to aging, improper repair, rehabilitation, or lack of proper maintenance. There is a wide research gap in bridge asset management particularly in the field of proper structural evaluation of bridges. Extending the useful service life of aging bridges is very important for the transportation industry. One of the problems faced by the US transportation industry is the degradation of structural components of highway bridges (normal deterioration and natural disasters like Hurricane Sandy etc.). The main reason behind these problems is the over usage of highway bridges beyond their useful service life in association with improper bridge asset management. The structural evaluation of the bridges is mostly done visually which varies in interpretation based on the judgement of the inspector (structure evaluator). In degradation model analysis, failure can be directly related to a change over time in a measurable structural parameter. This opens up the possibility of measuring degradation over time and using those data to extrapolate when failure will occur. This approach would allow us to fit acceleration models and life distribution models without actually waiting for failures to occur. In this analysis, various structural parameters, which drift monotonically (upwards, or downwards), can be measured over time towards a specified critical value corresponding to their failure stage. The aim is to fit models using degradation data instead of failures. The degradation model will be multidimensional, incorporating infrastructure element type, exposure, and time.
展开▼
