Scour Prediction at Bridge Piers in Cohesive Bed Using Gene Expression Programming

Accurate and reliable estimation of the scour depth at a bridge pier is essential for the safe and economical design of the bridge foundation. The phenomenon of scour at the pier placed on sediments is extremely complex in nature. Only a limited number of studies have been reported on local scour around bridge piers in cohesive sediment mainly due to the fact that scour modeling in cohesive beds is relatively more complex than that in sandy beds. Recent research has made good progress in the development of data-driven technique based on artificial intelligence (AI). It has been reported that AI-based inductive modeling techniques are frequently used to model complex process due to their powerful and non-linear model structures and their increased capabilities to capture the cause and effect relationship of such complex processes. Gene Expression Programming (GEP) is one of the AI techniques that have emerged as a powerful tool in modeling complex phenomenon into simpler chromosomal architecture. This technique has been proved to be more accurate and much simpler than other AI tools. In the present study, an attempt has been made to implement GEP for the development of scour depth prediction model at bridge piers in cohesive sediments using laboratory data available in literature. The present study reveals that the performance of GEP is better than nonlinear regression model for the prediction of scour depth at piers in cohesive beds.