On the performance of new local search heuristics for annual crop planning: Case study of the Vaalharts irrigation scheme
Journal of Experimental and Theoretical Artificial Intelligence
School of Mathematics, Statistics and Computer Science, University of Kwa-Zulu Natal, University Road, Private Bag X 54001, Westville, Durban, South Africa
This paper investigates the capabilities of three new local search (LS) metaheuristic algorithms in determining solutions to an annual crop planning (ACP) problem at an existing Irrigation Scheme. ACP is an optimisation problem in agricultural planning which involves determining resource allocation solutions amongst the various crops that are required to be grown at an irrigation scheme, within a year. The LS algorithms investigated are the best performance algorithm (BPA), the iterative best performance algorithm (IBPA) and the largest absolute difference algorithm (LADA). To determine the relative merits of the solutions found by these algorithms, their solutions have been compared against the solutions of two well-known LS metaheuristic algorithms and four population-based metaheuristic algorithms in the literature. The results show that BPA, IBPA and LADA were competitive in determining solutions for this particular optimisation problem. © 2014 Taylor & Francis.
Crops; Heuristic algorithms; Irrigation; Iterative methods; Local search (optimization); Optimization; Stochastic systems; Absolute difference; Crop planning; Meta heuristics; Optimisations; Stochastic local searches; Algorithms
UKZN, Inyuvesi Yakwazulu-Natali