Fig. 5

Sketch of key characteristics of a dosing error distribution. The \(t_{central}\) and \(\sigma \) are derived from the first and second moment, respectively, calculated directly in the Z-domain, without the need for an inverse Z-transform, as is explained in the text. a Push-out effect before the tip of Poiseuille profile has reached the patient. b Decaying remnant of push-out effect, due to Poiseuille mixing. c Dosing error due to mechanical compliances