Sensitivity analysis is essential for assessing how input uncertainties impact atmospheric dispersion models, particularly under imprecision and vagueness. This study presents an advanced framework that applies intelligent Monte Carlo techniques to the Unified Danish Eulerian Model (UNI-DEM) within fuzzy and intuitionistic fuzzy contexts. Traditional Monte Carlo methods often struggle with high computational demands and limited ability to address fuzzy uncertainty. To overcome these challenges, we employ optimized stochastic strategies–lattice rules, stratified sampling, and a novel Intuitionistic Fuzzy Monte Carlo for SA (IFMCSA) approach. These techniques enhance efficiency while maintaining accuracy with imprecise input data. Our framework effectively quantifies the influence of model parameters on pollutant dispersion, reduces variance in sensitivity indices, and strengthens model robustness under uncertainty. The results confirm that intelligent Monte Carlo methods for SA, especially IFMCSA, are scalable tools for uncertainty quantification in environmental systems.