This article presents a method based on evolutionary algorithms to optimize an automated irrigation system that accounts for multiple factors relevant to efficient water management. The proposed solution models various types of sprinklers and adapts to parameters such as soil absorption capacity and the specific water requirements of each crop. Using a vector representation to determine sprinkler positions and operating times, the method aims to reduce both over- and under-irrigation. System performance is evaluated through fitness functions incorporating application efficiency and water balance, while also minimizing the number of sprinklers required to achieve the target irrigation conditions. Results show that the proposed approach consistently outperforms the greedy baseline, achieving comparable irrigation deficits with costs reduced by more than 50%. Moreover, the proposed evolutionary algorithm identifies compromise solutions that extend the range of trade-offs beyond those attainable with baseline heuristic methods.