Urban green spaces provide vital ecological benefits, helping mitigate the urban heat island effect and improve living comfort. Traditional planning methods, however, often rely on subjective judgment and lack systematic optimization for complex, multi-objective constraints. This study integrates intelligent algorithms with environmental design, proposing a green space planning framework based on a multi-objective reinforcement-genetic hybrid algorithm, supported by GIS spatial analysis and ANN-based benefit prediction, forming an “evaluation–search–feedback” loop. The model is tested across different urban scales and functional zones, showing advantages in enhancing green space connectivity, reducing costs, and accelerating decision-making. A visual dashboard enables planners to adjust priorities in real time and track optimization processes. The research highlights the synergy between data-driven computation and design intent, offering a replicable approach for green infrastructure in dense urban areas.
Jin et al. (2025) studied this question.