While simplified computational thermal manikins (CTMs) are widely employed in CFD models of occupied indoor spaces in order to save the computational cost, a criterion of simplification is still absent and the effects of CTM simplification are yet not clear. In this study, six CTMs including a 3D scanned CTM and five simplified CTMs generated from various simplification approaches were employed to analyse the impact of CTM simplification on the prediction of airflow field and contaminant transport. Comparison of the predicted airflow field against the published data in the literature demonstrated that CTM simplification has a strong effect on the thermal airflow field prediction in the vicinity of manikin surfaces. For densely occupied indoor spaces such as a train cabin, the error induced by CTM simplification could be enlarged and further cause significant global error to the prediction of contaminant transport. This is especially true when contaminants are released from the CTMs. This study demonstrated that the mesh decimating algorithm is promising to simply CTMs that is not only able to reduce considerable computational cost but capable of maintaining an acceptable predictive error. (C) 2016 Elsevier B.V. All rights reserved.