The invention relates to a prediction method of organic compound blast characteristic based on genetic algorithm, and the method is determined by the molecular structure of organic compounds based on various blast characteristics of the organic compounds, and the molecular structure can be described by various parameters which reflect the characteristics of the molecular structure, namely, the organic compound blast characteristic can be expressed by the function of chemical construction parameters. The method firstly calculates structure parameters used for reflecting various molecular structure information according to the molecular structure of the organic compound, thus realizing parametric description of the molecular structure information; secondly, the genetic algorithm is applied to carry out characteristic variables screening, a group of parameters, which are taken as the descriptors of the molecular structure and closely related to the related blast characteristics and contain abundant structure information, are screened from a great amount of calculated structure parameters. Based on the above steps, an appropriate statistical modeling method is adopted to carry out statistical learning for the intrinsic quantitative relation between the selected descriptors and the related blast characteristics, thus obtaining a quantitative function model between the molecular structure and the related blast characteristics. The method has the advantages of simpleness, high prediction accuracy and providing a rapid, novel and accurate prediction method for the organic compound blast characteristics.