IGD values in robustness testing.

The acceleration of global urbanization and the rapid growth of urban populations have intensified the complexity and urgency of parking demand. In megacities with limited land resources, efficiently addressing diverse parking needs has become a critical issue for sustainable urban development. Multi-objective optimization methods are widely applied to tackle such challenges, providing decision-makers with a set of optimal solutions that balance multiple objectives. However, existing studies often lack quantitative analyses of the relationships among these solutions, limiting their applicability in accommodating decision-makers with varying preferences. This study focuses on Jing’an District in Shanghai, a representative region of a Chinese megacity, to address this global issue. Based on real-world data, a multi-objective optimization model is constructed considering convenience, coverage, and cost-efficiency. The model is solved using an improved Non-dominated Sorting Genetic Algorithm II (NSGA-II), which dynamically adjusts crossover and mutation rates. Furthermore, the Pareto solution set is quantitatively analyzed from a cost-benefit perspective by integrating marginal benefit theory. This approach provides robust support for decision-makers seeking an optimal balance between cost and benefit, offering scenario-specific strategies. The findings of this study not only present an innovative, systematic, and flexible solution to the “parking dilemma” in high-density residential areas but also provide practical guidance and insights for other large cities in the planning and implementation of smart underground parking facilities.