A Generalized Golden Rule Representative Value for Multiple-Criteria Decision Analysis

Multicriteria decision analysis evaluates multiple conflicting criteria in making, but are typical evaluating options. As the existing ordering operations involved multicriteria making cannot easily be implemented with intervals, we assume that scalar representative values intervals can effectively avoid this issue. To deal interval-valued criteria, propose a generalized golden rule value approach, which involves sigmoid function of backpropagation neural networks to tune parameters. Our approach considers uncertainties and side effects interval variables improve individual values. Based on numerical examples, address effectiveness proposed provide specific application concerning satisfaction.