计算指标权重。判断矩阵构建之后，先将矩阵每行的值分别进行相乘得到Ｑｉ，

Calculate indicator weights. After the judgment matrix is ​​constructed, first multiply the values ​​of each row of the matrix to obtain Qi, and then open n to the power of Qi to obtain W, where n is the order of the matrix, and then normalize n W to obtain The number is the weight value of the corresponding indicator. ④Check whether the established judgment matrix complies with the principle of consistency. Check the consistency of the judgment matrix. In order to ensure the coordination of the judgment matrix, a one-time inspection must be performed. First calculate the maximum eigenvalue, and then calculate the consistency index CI. Since some random factors may cause inconsistency with the principle of consistency, it is necessary to refer to the random consistency index RI when checking the established matrix. The value of RI can be obtained by referring to the table, and the possible random factors can be eliminated to make the test result more accurate. Use RI to measure CI and get a test coefficient CR, CR=＾, the judgment method is, if CR<0.1 , Then the established judgment matrix conforms to the principle of consistency, if not, then the established judgment matrix has not passed the test, and a new judgment matrix needs to be constructed. There is a one-to-one correspondence between the value of RI and the order of the judgment matrix, as shown in the following table.

Calculate the metric weight. After judging the matrix construction, the values of each row of the matrix are multiplied to get Qi, then Qi is open n times to get W, n is the order of the matrix, and then n W is normalized, the resulting number is the weight value of the corresponding indicator. (4) To test whether the judgment matrix established conforms to the principle of consistency. The judgment matrix is tested for consistency. In order to ensure the coordination of the judgment matrix, a one-time test must be carried out. First of all, the maximum characteristic value is calculated, and then the consistency indicator CI is calculated, because some random factors may cause non-conformity principle, so to test the established matrix, we need to refer to the value of the random consistency indicator RI, the value of RI can be obtained by reference to the table, cut out the possible random factors, so that the test results are more accurate, to MEASURE CI, to get a test coefficient CR, CR, the judgment method is, if CR 0. 1, then the established judgment matrix conforms to the principle of consistency, if not, then the established judgment matrix has not passed the test, need to build a new judgment matrix. The value of RI has a one-to-one relationship with the order of the judging matrix, as shown in the following table.

Calculate the index weight. After the judgment matrix is constructed, the values of each row of the matrix are multiplied to get Qi, and then Qi is opened to the nth power to get W. n is the order of the matrix, and then n w are normalized, and the number obtained is the weight value of the corresponding index. ④ Test whether the judgment matrix is consistent with the principle. Check the consistency of the judgment matrix. In order to ensure the coordination of the judgment matrix, a one-time test must be carried out. First of all, we need to calculate the maximum eigenvalue, and then calculate the consistency index CI. Because some random factors may cause inconsistency, we need to refer to the value of the random consistency index RI when we want to test the established matrix. The value of RI can be obtained from the reference table, and the possible random factors can be removed to make the test results more accurate I to measure Ci, and get a test coefficient Cr, Cr = ＾. The judgment method is: if Cr ＜ 0.1, then the established judgment matrix conforms to the consistency principle; if not, then the established judgment matrix does not pass the test, and a new judgment matrix needs to be constructed. There is a one-to-one correspondence between the value of RI and the order of the judgment matrix, as shown in the table below.<br>