基于TOPSIS的改进云PDR多准则决策方法

Improved Cloud PDR Multiple Criteria Decision Making Method Based on TOPSIS

为解决多准则决策在考虑数据波动性与不确定性特征时信息利用不充分、主观性强等问题,在云PDR方法基础上引入了TOPSIS中正负理想解、贴近度等概念,提出改进的云PDR多准则决策方法。首先,在属性值云转换基础上生成正负理想云,依照云PDR方法计算云优势度、α概率云优势类和云优势度矩阵,得到全部方案的综合云优势度;其次,基于模拟退火算法中Metropolis准则对综合云优势度做加权向量化处理,并利用改进的距离公式计算各方案与最优、最劣方案间的距离;再次,将距离转化为贴近度对方案排序;最后,结合算例验证了改进云PDR方法的可行性和有效性,为多准则决策提供了新的技术与方法。

In order to solve the problem of insufficient information utilization and strong subjectivity when considering the characteristics of data volatility and uncertainty in multiple criteria decision making,the concept of TOPSIS positive and negative ideal solution and closeness is introduced on the basis of cloud PDR multi-criteria decision-making method.First,generate positive and negative ideal clouds on the basis of attribute value cloud conversion.Then,according to the cloud PDR method,calculate the cloud dominance,α probabilistic cloud dominance class and cloud dominance matrix in order to obtain the comprehensive cloud dominance of all schemes;Secondly,the comprehensive cloud dominance is weighted and vectorized based on the Metropolis criterion in the simulated annealing algorithm Process,and use the improved distance formula to calculate the distance between each plan and the best and worst plan;Again,transform the distance into closeness to sort the plans;Finally,combine the calculation examples to verify the feasibility and effectiveness of the method,provides a new technical path for multiple criteria decision making.