Atherosclerotic Cardiovascular disease is the leading cause of myocardial infarction, causing nearly 1 in 4 deaths globally. Current treatment relies on statins, like simvastatin, which can reduce disease progression, but their use often leads to side effects and is limited in effectiveness. pH- and reactive oxygen species (ROS)-responsive nanoparticles offer a promising strategy for targeted drug delivery; however, optimal design parameters are not well established. This study developed a transport-aware computational framework that combines dimensionally consistentpH/ROS release kinetics with a multi-compartment pharmacokinetic model, Pareto screening, uncertainty quantification, and global sensitivity analysis. Nanoparticle-bound and free drug were simulated in blood, the mononuclear phagocyte system (MPS), healthy arterial wall, plaque cap, and plaque core. A reduced ROS-release model predicted a fully withheld 5 mM hydrogen peroxide curve with R² = 0.898. Screening 3,000 designs per mechanism identified a dual-responsive design that increased normalized plaque-free drug AUC by 39% relative to ROS-only nanoparticles (1.60versus 1.15), accompanied by a 17% increase in off-target AUC (3.54 versus 3.02). Across 5,000 paired uncertainty draws, dual responsiveness produced greater plaque exposure in 97.0% of simulations and a higher stimulus enhancement ratio in 64.1%. Plaque pH, carrier pH50, and baseline leakage were the stronges determinants of selectivity. The framework supports rational design by identifying robust exposure gains, clarifying formulation tradeoffs, and prioritizing experiments that can strengthen predictive confidence.