ARTIFICIAL INTELLIGENCE IN AGRICULTURE 4.0: ANT COLONY APPLIED TO WATER AND NITROGEN OPTIMIZATION
Keywords:
Productive Efficiency, Agricultural Resource Management, Production Functions, Smart Agriculture, Bio-Inspired Optimization, Decision Support SystemsAbstract
This study investigates Ant Colony Optimization (ACO) as an Artificial Intelligence strategy for the joint economic tuning of water and nitrogen in iceberg lettuce and melon crops, within the Agriculture 4.0 framework. Well-established agronomic production functions from the literature are replicated and embedded in a transparent two-dimensional grid formulation, in which pheromone deposition is proportional to performance and global evaporation controls the balance between exploration and intensification. The analysis focuses on solution quality, convergence dynamics, and sensitivity to key algorithm parameters and grid resolution. The economic optima obtained with ACO are consistent with results reported for existing AI/optimization-based tools and methods (such as INTELIAGRI, MBL, and Pattern Search), including a case in which the optimum lies on the boundary of the decision domain. Net revenues are computed under a uniform economic scenario, ensuring comparability across crops and approaches. The paper concludes by summarizing practical configuration guidelines for ACO and discussing their implications for the design of decision support systems in Agriculture 4.0 aimed at efficient water and nitrogen management.
