Crop simulation models are an effective and important application widely used in crop yield research and prediction, and the correction and optimization of the parameters are the premise of model simulation. The common trialerror method is lack of objectivity and of low efficiency. A sensitivity analysis of input parameters of crop models is carried out to identify the sensitivity of the parameters in simulation process, which can effectively identify the key parameters and reduce the number of parameters, so as to lay a foundation for the correction and optimization of the parameters. This paper discusses the advantages and disadvantages of both local sensitivity and global sensitivity methods of crop model analysis, such as the Fourier amplitude sensitivity test (FAST), Morris, LHOAT, Generalized Likelihood Uncertainty Estimation (GLUE), Sobol and extended Fourier amplitude sensitivity test (EFAST). The current status of research methods in crop models is reviewed and the prospects of sensitivity analysis of crop model parameters are put forward.