摘要
针对权系数信息不完全确定且方案的准则值为确定语言等级或位于两个连续语言等级之间或语言区间,甚至缺失的群决策问题,提出一种群体语言决策方法。该方法将各准则的语言值转换成二元语义,利用WC-OWA和OWA算子进行准则集结和方案集结,通过方案贴近度和决策者给出的准则权系数的不完全确定信息,建立非线性规划模型,利用遗传算法对模型进行求解得到准则的最优权系数,进而得到方案的群体集成值,并将其转化为二元语义,利用二元语义的有序性得到方案的排序。实例计算表明了该方法的可行性和有效性。
For the group decision-making problems in which criteria' s weights are incompletely certain, and criteria values of alternatives are represented with the form of linguistic grade , or between two linguistic grades, or linguistic interval, or even default, a new method based on WC-OWA and two-tuple is proposed,. In this method, linguistic values of all criteria are transformed into two-tuple, criteria values are aggregated with WC-OWA operator, and alternatives are aggregated with OWA operator. A nonlinear programming model is constructed based on close degree alternatives and incompletely certain information on weights. The optimal criteria' s weights are obtained by using ge- netic algorithms to solve the nonlinear programming model, and further the aggregative values on alternatives are worked out and transformed into two-tuple. The rank of the alternatives is obtained by using order property of two-tuple. An example shows the feasibility and effectiveness of the proposed method.
出处
《系统工程学报》
CSCD
北大核心
2010年第2期190-195,共6页
Journal of Systems Engineering
基金
国家自然科学基金资助项目(70771115)
国家自然科学基金创新群体资助项目(70921001)
国家自然科学基金重点资助项目(70631004)
关键词
多准则决策
群决策
语言评价
语言区间
二元语义
multi-criteria decision-making
group decision-making
linguistic assessment
linguistic interval
two-tuple linguistic