Joint Irrigation and Fertilization Scheduling in Multi-Crop Agricultural Systems Using the Gliding Snake Optimizer

Abstract

Efficient irrigation and fertilization scheduling for multi-crop farmlands represents a critical challenge in modern precision agriculture, particularly under increasing water scarcity and food security pressures. This paper proposes the application of the Gliding Snake Optimizer (GSO), a recently introduced bio-inspired meta-heuristic algorithm, to solve the combined crop irrigation and fertilization scheduling problem. The proposed model encodes water volumes, nutrient dosages (N, P, K), and event timing into a unified decision vector and optimizes a composite objective function that simultaneously maximizes normalized crop yield while minimizing water cost, fertilizer expenditure, and phonological timing penalties. Seven agronomic and resource constraints are enforced via an adaptive penalty mechanism. Extensive experiments on a benchmark scenario comprising four representative crops (wheat, corn, tomato, olive) demonstrate that GSO achieves statistically superior performance compared to four established algorithms -- Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA), and Grey Wolf Optimizer (GWO) -- across all evaluation metrics. GSO reduces water consumption by up to 23.4% and improves simulated yield by 18.7% relative to a conventional scheduling baseline. Wilcoxon signed-rank tests confirm that performance differences are statistically significant (p < 0.05) in 30 independent runs. The results demonstrate the strong suitability of GSO for constrained agricultural scheduling optimization problems.