In this paper, a method based on the artificial neural network technology, which is used to iden-tify two moving vehicles load on a simply supported bridge, is discussed. A multi-step identification procedure, which can be used effectively to identify multiple vehicle parameters such as loadlocation, load magnitude, andmoving speed, is proposed. The sensitivity of various input variables to the output and the effect of variation ofsample number to the identification result are investigated. The example data for training and verification aregenerated by theoretical simulation. The identification simulation shows that the proposed ANN method andprocedure can identify the moving vehicle load problem with acceptable accuracies. The ANN can also identifythe vehicle loads withclosed spacing and high moving speed, for which a conventional TDM or FTDM does notgive good identification results.