根据质量管理体系运行有效性模糊综合评价矩阵,从底层指标隶属度出发到求出体系运行有效性隶属度,要经过一系列的隶属度转换;但是,已有的隶属度转换方法不能揭示指标隶属度中“哪部分对目标分类有用,哪部分无用”;结果是指标隶属度中原本对目标分类不起作用的冗余数值也被用于计算目标隶属度。为此,从对目标分类角度设计一种滤波器,能识别指标隶属度中对目标分类不起作用的冗余数值,提取对目标分类起作用的有效值用于计算目标隶属度。由此建立实现隶属度转换的一般算法,并用于质量管理体系运行有效性评价。
The fuzzy evaluation matrix of the effectiveness of QMS ( Quality Management System) is one hierarchical structure and the decision process of this system actually does a series of calculation to synthesis the membership degree from bottom indexes to objective. But, existing converse algorithms of membership degrees are questionable if distinguishing the useful or useless part of membership degrees of index so that redundant data, which are useless to objective classification, are involved in calculation of membership degrees of the objective. Therefore, based on the pure synthesis of membership degree, a filtering algorithm, which can identify and filter the redundant membership degrees of index and redundant data useless for objective classification, is proposed. Valid values, which are useful for objective classification, are filtered out to synthesize exactly to the memberships degree from indexes to objective. Based on this, a new method to evaluation on the effectiveness of QMS is proposed.