Optimization of Cropping Patterns Using Elitist-Jaya and Elitist-TLBO Algorithms

Abstract

The Karjan irrigation project of Gujarat in Western India faces an acute shortage of irrigation water during the non-monsoon seasons. Improper water management in the area may result in reduced yields and hence low net incomes to the farmers. Optimal cropping pattern models were developed under different constrained environments using the improved versions of Jaya algorithm (JA) and teaching learning based optimization (TLBO) algorithm by incorporating the elitist concept, namely EJA and ETLBO, to maximize the net annual benefits. The advantages of EJA and ETLBO are that they do not require any algorithm-specific parameters and only need common controlling parameters, such as population size and number of iterations. Two different models of maximum and average cropping patterns were developed. Different elite sizes were tested with various combinations. The results of EJA and ETLBO were compared, and whether the improved version of the algorithm will enhance the results was checked. Moreover, the findings were compared with those of the linear programming (LP) model. It was observed that maximum net benefits were obtained by EJA for both the models. The results demonstrate a substantial gain in the cultivation of banana, cotton, sugarcane, and groundnuts. Based on the results, it is concluded that EJA outperforms ETLBO, JA, TLBO, and LP.