广东、广西、浙江、福建、湖南、河南等省(区),都应用马尔可夫链方法进行林业害虫的预测预报,但在预测未来年份马尾松毛虫的发生面积、划分等级和确定状态范畴时,常受人为的主观意识影响。由于分级标准不同,所得到的测报结果出现了较大的差异。为了减少测报误差,更准确地反映马尾松毛虫发生的实际面积,现将优选法改进分级标准的研究综述如下。 The prediction and prediction of forestry pests are carried out by Markov chain method in Guangdong, Guangxi, Zhejiang, Fujian, Hunan and Henan provinces. However, in predicting the occurrence area, classification level and determining the state category of Dendrolimus punctatus in the coming years, , Often affected by subjective human consciousness. Due to different grading standards, the obtained results of the survey showed a big difference. In order to reduce the measurement error and reflect the actual area of ​​Dendrolimus punctatus more accurately, the research on the grading standards of optimization method is summarized as follows.