Chronic exposure to inorganic arsenic (iAs) in the human population is associated with various internal cancers and other adverse outcomes. The purpose of this study was to estimate a population-scale exposure risk attributable to iAs consumptions by linking a stochastic physiological-based pharmacokinetic (PBPK) model and biomonitoring data of iAs in urine. The urinary As concentrations were obtained from a total of 1,043 subjects living in an industrial area of Taiwan. The results showed that the study subjects had an iAs exposure risk of 27 % (the daily iAs intake for 27 % study subjects exceeded the WHO-recommended value, 2.1 μg iAs day(-1) kg(-1) body weight). Moreover, drinking water and cooked rice contributed to the iAs exposure risk by 10 and 41 %, respectively. The predicted risks in the current study were 4.82, 27.21, 34.69, and 64.17 %, respectively, among the mid-range of Mexico, Taiwan (this study), Korea, and Bangladesh reported in the literature. In conclusion, we developed a population-scale-based risk model that covered the broad range of iAS exposure by integrating stochastic PBPK modeling and reverse dosimetry to generate probabilistic distribution of As intake corresponding to urinary As measured from the cohort study. The model can also be updated as new urinary As information becomes available.