Aiming at the problem of long time of vehicle engine reliability life test,the grey system forecasting theory is used to forecast vehicle engine reliability life test data,and the unbiased grey GM（ 1,1） power model for engine reliability life test data is proposed. By examining the overall distribution function equality of test data sequence and forecasted data sequence,it is confirmed that the proposed unbiased grey GM（ 1,1） power model can be effectively used for the prediction of engine reliability life test data. The example results show that the grey theory forecasting method for the prediction of engine reliability life test data has high precision,and distribution function parameters of test data sequence and forecasted data sequence are very close. Using the proposed method can shorten engine reliability life test time and save test cost.