A mathematical lumped parameter model of the oscillometric technique for indirect blood pressure measurement is presented. The model includes cuff compliance, pressure transmission from the cuff to the brachial artery through the soft tissue of the arm, and the biomechanics of the brachial artery both at positive and negative transmural pressure values. The main aspects of oscillometry are simulated i.e., the increase in cuff pressure pulsatility during cuff deflation maneuvers, the existence of a point of maximum pulsations (about 1.5 mmHg) at a cuff pressure close to mean arterial pressure, and the characteristic ratios for cuff pressure pulsatility at systole and diastole (0.52 and 0.70, respectively, with this model, using basal parameters and an individual set of data for the arterial pressure waveform). Subsequently, the model is used to examine how alterations in some biomechanical factors may prejudice the accuracy of pressure measurement. Numerical simulations indicate that alterations in wall viscoelastic properties and in arterial pressure pulse amplitude may significantly affect the accuracy of pressure estimates, leading to errors as great as 15-20% in the computation of diastolic and systolic arterial pressure. By contrast, changes in arterial pressure mean value and cuff compliance do not seem to have significant influence on the measurement. Evaluation of mean arterial pressure through a characteristic ratio is not robust and may lead to misleading results. Mean arterial pressure may be better evaluated as the lowest pressure at which cuff pulse amplitude reaches a plateau. The obtained results may help to explain the nature of errors which usually limit the reliability of arterial pressure measurement (for instance in the elderly).