Individualized models of respiratory mechanics may help to reduce potential harmful effects of ventilation therapy by predicting the outcome of certain ventilator settings. The underlying models are commonly identified by iterative error-mapping methods, such as the Levenberg-Marquardt Algorithm, requiring initial estimates for the patient specific parameters. The quality of the initial estimates has a significant influence on identification efficiency and results. An iterative integral-based parameter identification method was applied to a linear 2nd order respiratory mechanics model. The method was compared to the Levenberg-Marquardt Algorithm using clinical data from 13 Acute Respiratory Distress Syndrome (ARDS) patients. The Iterative Integral-Based Method converged to the Levenberg-Marquardt solution two times faster and was independent of initial estimates. These investigations reveal that the Iterative Integral-Based Method is beneficial with respect to computing time, operator independence and robustness.