Aim: This prospective study aims to investigate the factors influencing voriconazole trough concentration (Cmin), develop a population pharmacokinetics (PPK) model and recommend an appropriate voriconazole dosing regimen for children with hematological malignancies. Methods: Prospectively enrolled a total of 70 children aged <18 years and 149 samples. The factors influencing voriconazole Cmin were analyzed by univariate analysis and multiple linear regression analysis. Nonlinear mixed effects modeling (NONMEM) was applied to establish the PPK model. Dosage simulation based on albumin (ALB) levels and CYP2C19 genotype. Results: Multiple linear regression results demonstrated that route of administration, ALB and concomitant administration with glucocorticoid (GLU) and proton pump inhibitors (PPIs) were significant factors of voriconazole Cmin. A one-compartment model could best describe the pharmacokinetics of voriconazole. The extensive metabolizers (EM), ALB were significant covariates of clearance (CL). The typical value of CL, the volume of distribution (V) and oral bioavailability (F) were 1.52 L/h, 35.7 L and 0.909, respectively. The recommended dosing regimens for EM patients with ALB level of 20.0~35.0 g/L, 35.1~45.0 g/L and 45.1~55.0 g/L were4, 8 and 12 mg/kg intravenously or orally twice daily, respectively, and were 2, 4 and 7 mg/kg by intravenous or oral administration twice daily for non-EM. Conclusion: We found that route of administration, ALB and co-administration of GLU and PPI had quantitative relationships with voriconazole Cmin. The combination of CYP2C19 genotype and ALB levels to determine the initial dosing regimen of voriconazole could provide a reference for individualized treatment in children with hematological malignancies.