基于灰色预测GM(1,1)模型的火灾损失预测

Fire loss prediction based on gray prediction GM (1,1) model

基于灰色预测GM(1,1)模型,根据2002-2009年全国火灾损失数据,包括每年全国火灾发生起数、死亡人数、受伤人数、直接经济损失,预测2010-2012年全国火灾损失情况,并将灰色预测结果与2010年和2011年火灾损失数据进行对比。结果表明:全国火灾发生起数、死亡人数、受伤人数预测结果合格,且火灾发生起数、火灾死亡人数预测精度较高;每年全国火灾直接经济损失规律性很弱,用灰色预测GM(1,1)模型无法预测2010-2012年全国火灾直接经济损失。

The fire loss in 2010 to 2012 was predicted based on the gray prediction GM （1,1） model, combined with the fire loss data in China between 2002 and 2009, including the num- ber of fire, the number of death, the number of injury and the direct economic loss in fire per year. And the prediction results were compared with the actual fire loss in 2010 and 2011. The results demonstrated that the prediction results, for the number of fires, the number of death and the number of injury in fire, are eligible, and the predicted number of fire and the forecast number of death fit well with the actual number in 2010 and 2011. However, the direct economic loss in 2010 to 2012 could not be predicted by gray prediction GM（1,1） model, due to the bad regularity of the corresponding data.