Abstract: The applicability of moving load identification method is a crucial parameter, which can be used to determine whether the method can be adopted in the field or not. Based on the method of moments and the theory of moving load identification, an improved time domain method (ITDM) is proposed for the identification of moving vehicle loads. The moving loads are expressed in terms of a series of basic functions, such as orthogonal Legendre polynomials and Fourier series, and are estimated by solving the new system equations developed based on the basic functions. Some numerical simulations and experiments in laboratory were carried out, which aimed at conducting a comparative study and evaluating the effect of the basic function number and the CPU executive time on the ITDM. The illustrated results show that the ITDM almost has as good potentials as the time domain method (TDM), but it can save the CPU executive time as well as with a better flexibility and applicability when compared with the TDM, especially when the Fourier series are adopted as the basic functions. This is beneficial to the real time analysis of moving load identification in field.