基于组合权重的系统评价模型

提出了基于组合权重的系统评价新模型 ( CWSE) ,即 :直接根据评价指标样本数据集 ,用基于加速遗传算法的投影寻踪方法确定各评价指标的分类权重 ,用基于加速遗传算法的层次分析法确定各评价指标的排序权重 ,用加速遗传算法对各评价指标的分类权重和排序权重进行综合得到组合权重 ,然后以这些组合权重与各评价对象相应评价指标的标准化值进行加权平均 ,得到系统评价的综合指标值 ,据此可对各评价对象进行分类排序 .用 CW SE模型评价中国 30个区域 1995年开发度的结果表明 ,根据开发度的强弱可把这些区域分成 3个强开发区域、6个较强开发区域、10个中等开发区域和 11个弱开发区域 ;CWSE模型简便、通用 ,计算结果较为客观和稳定 ,为系统工程理论和实践提供了新的研究方法 .

System Evaluation Model Based on Combined Weights

A new system evaluation model, named CWSE, is presented based on combined weights. Directly utilizing samples of evaluation indexes, the classified weights of evaluation indexes can be estimated by using projection pursuit based on accelerating genetic algorithm developed by the authors, the sort order weights of evaluation indexes can be estimated by using the analytic hierarchy process based on accelerating genetic algorithm, the combined weights of evaluation indexes can be estimated by using accelerating genetic algorithm, and then the comprehensive index of system evaluation can be obtained according to the combined weights and standardization values of evaluation indexes. The research result of CWSE model applied to evaluating development degree of 30 regions in China shows that these regions can be divided to 3 strong development regions, 6 relatively strong development regions, 10 middle development regions, and 11 weakly development regions, that CWSE model is both simple and general, and that its computation result is objective and stability, which provides a new methodology to theory and practice of systems engineering.