基于目标导向的多属性决策属性价值公度方法

MACBETH方法在多个属性维度上确定出的一组属性Good值(或Neutral值)并不能真地具有保证属性价值公度的完全相同的偏好涵义.为克服该方法缺陷,基于包容性属性价值函数,通过引入数据包络分析技术和马尔可夫理论给出了以目标参照方案为导向的目标导向属性价值函数和多属性决策属性价值公度的方法程序.相比于具有较大随意性确定属性Good值和Neutral值的MACBETH方法,依赖目标导向属性价值函数的多属性决策属性价值公度程序具有更明显的科学合理性,因此具有发展MACBETH方法,推动实际决策应用该方法及相关多属性偏好关联决策模型的创新价值.基于决策实验的应用分析直接验证了多属性决策属性价值公度方法程序的科学有效性,间接验证了包容性属性价值函数和目标导向属性价值函数的科学理性.

A targets-oriented approach to commensurable preference-strength values in multiple attributes decision-making

The set of‘Good’(‘Neutral’) performance values given by the MACBETH approach to multiple attributes are not of the same preference meaning that guarantees the commensurability in decision-makers' preference-strength values (PSVs) on attribute performances. To solve this approach shortcoming, a targets-oriented PSV function (TOPSVF), which takes purposeful reference options as decision targets, and a methodological procedure for measuring the commensurable PSVs in multiple attribute decision-making, are presented by data envelopment analysis and Markov theory, and based on the inclusive PSV function (IPSVF). Compared with the MACBETH approach in which there is strong arbitrariness in the given‘Good’(‘Neutral’) performance values, the presented procedure dependent on the TOPSVF is of better science reasonability, and thus of innovation worth in improving the approach and promoting it as well as the multiple attribute decision-making preference dependence decision models related with it to be applied to real-world decisions. The application of the procedure to an experimental decision has directly verified its effectiveness and indirectly verified the science reasonability of the IPSVF and the TOPSVF.