摘要
为了减少决策信息损失,针对区间数决策问题,提出了一种立体决策方法。将m维区间数决策向量看作具有2m个顶点的超矩形。利用Ls(2m)正交表在超矩形上均匀、分散地选取s个顶点代表该超矩形,再根据不同超矩形对应顶点之间的相对熵度量超矩形之间的差异,最后利用各决策方案同理想解的贴近度进行决策方案排序。实例验证表明该文方法有效、可行。由于选取了多个多维实数点代表区间数决策向量,在丰富了决策信息的同时也增加了计算量。
To decrease the loss of decision making information,a stereo decision making method is proposed for the interval number decision making problem. An mdimension interval number decision making vector is regarded as a hyper rectangle with 2 m vertices. s vertices are selected equably and dispersedly from the hyper rectangle using an Ls （ 2^m ） orthogonal array to represent the hyper rectangle. The difference between two hyper rectangles is measured by the relative entropies between the corresponding vertices of the hyper rectangles. All decision alternatives are ranked using the degree of similarity between every decision alternative and ideal solution. An illustrative example is given to demonstrate the feasibility and validity of the proposed method. The decision making information and calculation is increased because of selecting many multi-dimension real numbers to represent interval number decision vectors.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2014年第6期823-828,共6页
Journal of Nanjing University of Science and Technology
基金
国家自然科学基金(71271114
71303004)
教育部人文社会科学规划基金(10YJA630020)
关键词
正交试验
区间数
正交表
超矩形
相对熵
贴近度
多维实数点
决策向量
orthogonal experiments
interval numbers
orthogonal arrays
hyper rectangles
relative entropies
similarity degrees
multi-dimension real numbers
decision vectors
