An arteriovenous fistula (AVF) is used as vascular access in hemodialysis. A previously developed patient-specific pulse wave propagation model [Huberts, 2011] might be useful to improve selection of the optimal AVF location (upper or lower arm) by predicting postoperative flow. However, personalizing the model parameters is difficult due to the limitations of measurement modalities, the large uncertainties in clinical measurements and the burden on the patient that should be minimized, resulting in sparse datasets. Therefore, it would be beneficial to identify the model parameters that are essential in the flow prediction and should thus be measured for each patient. Hence, the aim of this study was to identify the (non-) influential model parameters of the pulse wave propagation model applied to AVF surgery.