Repository Collections

Agriculture plays a central role in food security and rural livelihoods, yet traditional monocropping and intuition-based decision-making often lead to low productivity and inefficient resource use. Multicropping improves soil fertility, income diversification, and sustainability, but optimizing crop combinations across irrigation systems is a highly complex problem. Current studies are limited by the absence of benchmark datasets, single-crop focus, and reliance on isolated
algorithms. This paper surveys the role of Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Deep Learning (DL), and Reinforcement Learning (RL) in agricultural optimization, and proposes a hybrid framework that integrates GA, PSO, and DL as the core, with RL as a comparative adaptive layer. Selected algorithmic variants such as Adaptive GA, Multi-Objective PSO, LSTM/GRU, and Deep RL are identified as most suitable for multicropping. The proposed approach aims to generate synthetic datasets, optimize crop–irrigation strategies, and provide a decision support tool for farmers.