Purpose: The aim of this study was to build a population pharmacokinetics (PopPK) model of nalbuphine and to estimate the suitability of bodyweight or fixed dosage regimen. Method: Adult patients who were undergoing general anesthetic surgery using nalbuphine for induction of anesthesia were included. Plasma concentrations and covariates information were analyzed by non-linear mixed-effects modeling approach. Goodness-of-fit (GOF), non-parametric bootstrap, visual predictive check (VPC) and external evaluation were applied for the final PopPK model evaluation. Monte Carlo simulation was conducted to assess impact of covariates and dosage regimens on the plasma concentration to nalbuphine. Results: 47 patients aged 21-78 years with a body weight of 48-86 kg were included in the study. Among them, liver resection accounted for 14.8%, cholecystectomy for 12.8%, pancreatic resection for 36.2% and other surgeries for 36.2%. 353 samples from 27 patients were enrolled in model building group; 100 samples from 20 patients were enrolled in external validation group. The results of model evaluation showed that the pharmacokinetics of nalbuphine was adequately described by a two-compartment model. The hourly net fluid volume infused (HNF) was identified as a significant covariate about the intercompartmental clearance (Q) of nalbuphine with objective function value (OFV) decreasing by 9.643 (p < 0.005, df = 1). Simulation results demonstrated no need to adjust dosage based on HNF, and the biases of two dosage methods were less than 6%. The fixed dosage regimen had lower PK variability than the bodyweight regimen. Conclusion: A two-compartment PopPK model adequately described the concentration profile of nalbuphine intravenous injection for anesthesia induction. While HNF can affect the Q of nalbuphine, the magnitude of the effect was limited. Dosage adjustment based on HNF was not recommended. Furthermore, fixed dosage regimen might be better than body weight dosage regimen.