The aim of this study was to develop a population pharmacokinetic model of sufentanil and to assess the influence of covariates in critically ill children admitted to a pediatric intensive care unit. After institutional approval, 41 children were enrolled in the study. Blood samples for pharmacokinetic (PK) assessment were collected from routinely placed arterial catheters during and after discontinuation of infusion. Population nonlinear mixed‐effects modeling was performed using NONMEM. A 2‐compartment model described sufentanil PK sufficiently. Typical values of the central and peripheral volume of distribution and the metabolic and intercompartmental clearance for a theoretical patient weighing 70 kg were VC = 7.90 l, VT = 481 L, Cl = 45.3 L/h, and Q = 38.3 L/h, respectively. High interindividual variability of all PK parameters was noted. Allometric/isometric principles to scale sufentanil PK revealed that to achieve the same steady‐state sufentanil concentrations in plasma for pediatric patients of different body weights, the infusion rate should follow the formula (infusion rate for a 70‐kg adult patient, μg/h) × (body weight/70 kg)0.75. Severity of illness described by PRISM score, the monitored physiological and laboratory parameters, and coadministered drugs such as vasopressors were not found to be significant covariates.