With the rapid development of the global aviation industry, airport ground operations management is increasingly challenging. Ensuring safety, improving efficiency, and reducing environmental impacts constitute critical tasks. To address this, a mixed-integer linear programming model incorporating taxiway conflict prevention was developed. This model aimed to minimize taxi time and CO₂ emissions through dynamic optimization with the non-dominated sorting genetic algorithm II (NSGA-II). Implementation was conducted in Python for a major Chinese hub airport, with results compared against the commercial optimizer Gurobi. Computational findings reveal a 17.46% reduction in total taxi time and an 18.35% decrease in CO₂ emissions across 14 aircraft. The NSGA-II solution is found to be within 1.083% of Gurobi's optimal solution, while a 95.0% faster computation time is achieved. The capability of NSGA-II in handling large-scale multi-objective taxi path optimization problems is demonstrated. Operational efficiency is enhanced, and CO₂ emissions are significantly reduced by the proposed approach.