Prediction of emergency supplies demand based on Metabolism GM(1,1)++
When large outbreaks occur,the availability of emergency supplies is critical.Based on the metabolism GM(1,1)++model,this paper proposes a metabolism GM(1,1)++model,that is,new information is added,old information is removed from the original data,and data items with the largest relative residual difference between the original data and the fitting value are replaced by the fitting value.The method proposed in this paper can not only timely remove the old information with decreasing significance,but also add new information which can better reflect the current characteristics of the system,but also reduce the disturbance of other factors to the original data,so that the original data is more regular.In order to test the effectiveness of metabolism GM(1,1)++model,the prediction results were compared with those of traditional GM(1,1),new information GM(1,1)and metabolism GM(1,1)respectively.The experimental results show that the metabolism GM(1,1)++model has the smallest sum of squared errors,and the prediction accuracy is much better than the other three prediction models.
NETL
NSTL
万方数据
