World Journal of Modelling and Simulation

The movement of water into the soil under specific conditions is called infiltration. The study of infiltration is one of the most important necessities in irrigation and drainage engineering. Two popular infiltration models called Kostiakov and Modified Kostiakov, which are largely used in the fields, are considered for analysis. The problem is modeled as a least sum of squares of the differences in observed infiltration rates and computed infiltration rates over a specified time interval. In this paper, six optimization approaches are implemented to solve the models. They are (i) software Language for INteractive General Optimization (LINGO); (ii) Binary Genetic Algorithm (BGA); (iii) Hybrid Binary Genetic Algorithm (HBGA); (iv) Real Coded Genetic Algorithm using Laplace Crossover and Power Mutation (LX-PM); (v) its Hybrid version (H-LX-PM); and (vi) Random Search Technique (RST). The comparative performance of these methods is displayed for obtaining the infiltration parameters for the datasets (two each) of popular soils, namely Plain Field Sand (PFS) and Columbia Sandy Loam (CSL). To test model performance, the Nash-Sutcliffe efficiency is employed in this study.