ANFIS-based approach for scour depth prediction at piers in non-uniform sediments

An estimation of scour depth is a prerequisite for the efficient foundation design of important hydraulic structures such as bridge piers and abutments. Most of the scour depth prediction formulae available in the literature have been developed based on the analysis of the laboratory/field data using statistical methods such as the regression method (RM). Conventional statistical analysis is generally replaced in many fields of engineering by the alternative approach of artificial neural networks (ANN) and adaptive network-based fuzzy inference systems (ANFIS). These recent techniques have been reported to provide better solutions in cases where the available data is incomplete or ambiguous by nature. An attempt has been made to compare the performance of ANFIS over RM and ANN in modeling the depth of bridge pier scour in non-uniform sediments. It has been found that the ANFIS performed best amongst all these methods.