OBJECTIVE Linezolid can significantly impact drug-resistant tuberculosis (DR-TB) patient outcomes. However, the long-term use of this drug for TB treatment has been limited by adverse reactions and uncertainty regarding optimal dosage regimens for balancing drug efficacy and safety across different populations. This study attempted to find the optimal dosing regimen of linezolid in different populations. METHODS A total of 355 blood samples were collected from 126 DR-TB patients. Population pharmacokinetic analysis (using a one-compartment model) and dose simulations were conducted using NONMEM and R software. The ratio between the area under the free drug plasma concentration-time curve to the MIC (fAUC/MIC) of > 119 and trough concentration (Cmin) ≤ 2 mg/L served as efficacy and safety targets, respectively, toward the formulation of optimal dosage regimens based on a ≥ 90% cumulative fraction of response. RESULTS Body weight and blood urea nitrogen levels were the most significant covariates of apparent volume, while creatinine clearance and haemoglobin level significantly influenced apparent clearance. The probability of target attainment for different dosage regimens was evaluated via Monte Carlo simulation. For subjects with MICs of 0.125, 0.25 and 0.5 mg/L, specific total daily doses of ≥ 300 mg, ≥ 450 mg and ≥ 900 mg were required to reach the target, respectively. Subjects with body weight ≤ 70 kg and MIC ≥ 1 mg/L received a total 1200 mg daily dose to reach the probability of target attainment target. Notably, single dosing was safer than multiple dosing at the same daily dose. The optimal dosage regimens for subjects with body weight < 50 kg and ≥ 50 kg were 450 mg/d and 600 mg/d (once daily), respectively. CONCLUSIONS Optimal dosage regimens for patients weighing < 50 kg and ≥ 50 kg were 450 mg/d and 600 mg/d, respectively. A single dose was safer than multiple doses.