Abstract: Multiple-source information aggregation is a significant way to improve the reliability of the statistical data in the environment of natural disasters. However, the diversity of information sources often cause the inconformity and incompatible types of the aggregation information data which thereby produce isomerism data sequences that belong to different types of data. The modeling technology of grey system is applied to study the modeling method for grey isomerism data. First, the grey heterogeneous data is normalized based on the "Kernel" and "degree of greyness"; then a (1,1) DGM model of grey heterogeneous data "Kernel" sequence is established. According to the "Kernel" and the axiom of grey degree not-reducing, a grey prediction model of heterogeneous data is built by using the information field to which the maximum degree of greyness in the grey heterogeneous data sequence corresponding as the information field of the prediction results, Finally, the model is applied to forecast the demand volume of tent in the ya'an earthquake. The research results of the project will expand the traditional modeling objects of gray forecasting model from homogeneous data to heterogeneous data and has a vital significance for enriching and perfecting the grey prediction model theory system, and improving the efficiency of natural disasters and rescue.

Key words: grey theory, prediction model, grey isomerism data, emergency material demand