Selective Catalytic Reduction (SCR) systems play a crucial role in mitigating nitrogen oxides (NO \(_{\text {x}}\) ) emissions from combustion engines through the injection of a urea-based reductant into the exhaust stream. This research leverages extensive experimental data from real-engine tests to calibrate a 0-Dimensional model of the SCR system, trying to capture its physical and chemical dynamics as much as possible. The model utilizes a system of nonlinear differential equations to describe NH \(_3\) and NO \(_{\text {x}}\) concentrations, with parameters calibrated through a Particle Swarm Optimization (PSO) algorithm. Given the model’s complexity and the presence of experimental uncertainties, a statistical methodology is applied to produce mean values and confidence intervals (CIs) that account for experimental variability. The model’s predictions of NO \(_{\text {x}}\) emissions are presented as time-series data, with 95% CI bands covering around 90% of the experimental data. Future efforts will aim to enhance model calibration by ensuring consistent parameter sets across different experiments.