Combined Evaluation Method And Its Application

In the multi-attribute comprehensive evaluation, due to the uncertainty of decision-making, the difference of evaluation methods, a variety of viable evaluation methods for the same evaluation target get different evaluation results, that is, " Inconsistency among the evaluation conclusions drawn by different evaluation methods ", in order to resolve this problem, some scholars put forward the combination of thinking.No doubt, the combination evaluation can reduce the random error and systematic error of the evaluation process and improve the credibility of the results of the evaluation. Combination evaluation can be divided into soft,and hard direction.The former refers to different conclusions or conflict within the group to carry out consultations on the existence of differences of opinion, reflects the "large number of people satisfied" with the "reason" thinking; the latter refers to the combination process does not have subjective intervention , only through the automatic completion of the model, each evaluation conclusion will be fair and therefore a combination of evaluation findings has a strong objectivity. In accordance with the time involved in the model, the valuation methods can be divided into " the weight combination before reaching evaluation conclusions " and "conclusions of the evaluation findings ".In this paper, the theoretical study is mainly about inconsistency among the evaluation conclusions drawn by different evaluation methods. The results are as follows:(1) On the basis of the smallest deviation model and the entropy method, and also defining distance entropy, an objective combination evaluation method based on the distance and relative entropy was given.(2) On the basis of previous studies, and taking into account the defect of the average standard, a method based on cooperation game and the overall diversity was given to avoid the shortcomings of the average standard by introducting overall diversity model. Finally, numerical example showed the feasibility and effectiveness.(3) According to research results of the combination forecast, and expanding the variance-covariance model, we regarded the smallest deviation as the smallest error, and whole difference as the largest variance, An method based on the smallest deviation and the largest whole difference was given in conjunction with Jaynes maximum entropy model.Finally, numerical example show the feasibility of the method.(4) A combination evaluation method based on fuzzy clustering and interval number was given by extending the fuzzy C-means clustering from the point set to interval fuzzy C-means clustering and combinating with interval number sorting method. (5) A combination evaluation method based on the commonly used statistical models was given, the methods is simple and worth promoting.The applied research of the combination evaluation is about combination evaluation of lead-zinc industries sustainable development. The results are as follows:(1) Sustainable development mining indicators was established by collecting non-ferrous industry, mining indicators of sustainable development at home and abroad. Then China's lead and zinc sustainable development indicators was established through a questionnaire survey and indicator collection.(2) Evaluated economic, social, and environmental systems of the lead and zinc industry by using simple linear weighting, the entropy method, the ideal solution method and principal component analysis after collecting the actual data from year 1998 to 2006. Then evaluated single evaluation method by using distance entropy method and cooperative game and whole-diversity combination evaluation, the results showed that the two combination evaluation methods effective and feasible.