Estimation of Optimal Crop Plan Using Nature Inspired Metaheuristics

Irrigation management has gained significance due to growing social needs and increasing command for food grains while the available resources have remained limited and scarce. Irrigation management includes optimal allocation of water for irrigation purposes, optimal cropping pattern for a given land area and water availabilities with an objective to maximize economic returns. In the present study we consider an optimization model based on linear programming for determining optimal crop plan for command area of Pamba-Achankovil-Vaippar (PAV) link project, Kerala, India. The crop planning model considers various resource constraints (land area, seeds, manure, fertilizers etc.) availability etc. adaptive to national conditions, with the objective to maximize net irrigation benefits. For crop planning, the extent of quantity available for fertilizers, manure and seeds as inputs were unknown. Estimates for the extent of unknown minimum quantities of these resource inputs available are obtained with the help of crop planning model itself. For optimal releases made from reservoir using a multi-reservoir operation model, optimal crop plans are developed under adequate, normal and limited irrigation water defined by 50 percent, 75 percent and 90 percent water year dependable flows, respectively. The optimization model is solved using four popular Evolutionary Algorithms (EA) viz. Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Differential Evolution (DE) and Evolutionary Programming (EP). EA are compared with each other in terms of average CPU time, average number of generations, standard deviation etc. the algorithms are also compared with LINGO, a popular software used for solving LPP models.