Objective The importance of physiologically based harmacokinetic (PBPK) models in drug research and development is increasingly recognized but only a small number of PBPK models have been developed for botanical drugs to-date. The purpose of our study was to use green tea as an example to illustrate the development and validation of PBPK models for herbal products. Method:The green tea model consisted of a 5-compartment caffeine sub-model and three 11-compartment tea catechin sub-models. The caffeine sub-model was adapted from Ginsberg et a/. (J. Toxicol. Environ. Health 67: 297, 2004) and the catechin sub-models were extrapolated from the corresponding rat models developed in our laboratory. The caffeine and catechin sub-models were joined together at the liver compartment with the assumption that little or no pharmacokinetic interaction occurred between these chemicals. Model performance was assessed by comparing simulated caffeine/catechin plasma concentration-time profiles with experimental data from humans after consuming green tea or Polyphenon E (decaffeinated green tea). Result: The green tea model was able to provide an accurate description of the caffeine/catechin concentration-time profiles in the plasma Conclusion: The PBPK modeling approach as described in the present study potentially can be used to predict the pharmacokinetics of individual Chinese herbs and their combination in humans.