Machine Learning Solutions for Bridge Scour Forecast Based on Monitoring Data

Bridge scour is a challenge throughout the U.S.A. and other countries. Despite scale of issue, there still substantial lack robust methods for prediction to support reliable, risk-based management decision making. Throughout past decade, use real-time monitoring systems has gained increasing interest among state departments transportation across This paper introduces three distinct methodologies using advanced artificial intelligence (AI)/machine learning (ML) techniques based on data. Scour data included riverbed river stage elevation time series at bridge piers gathered from various sources. Deep algorithms showed promising in bed water level variations as early week advance. Ensemble neural networks proved successful predicting maximum upcoming depth, observed sensor onset episode, pier, flow characteristics. In addition, two common empirical models were calibrated Bayesian inference method, showing significant improvement accuracy. Overall, this novel approach risk by integrating emerging AI/ML with forecast.