GNMM and accurate longitudinal dispersion coefficient prediction

Longitudinal dispersion coefficient is a key variable for the description of the longitudinal transport in a river. In recent years, Artificial Neural Networks (ANNs) have become popular and useful tools for environmental modellers as they are perceived to overcome some of the difficulties associated with traditional statistical approaches. In these applications, however, little attention was given to the task of selection of the most appropriate ANN inputs. A Genetic Algorithm (GA) based variable selection technique has been proposed in our previous works (Genetic Neural Mathematical Method, hereafter called GNMM). The aim of the current chapter is to provide an insightful analysis into the technique that uses GAs as an ANN input variable optimization tool in the context of longitudinal dispersion coefficient prediction.