两级递阶软对策模型及其在冲突分析中的应用

Nigil Howard通过对亚对策的改进提出了软对策，但该理论仍基于效用进行分析。且没有涉及局中人个体偏好与整体偏好的不一致问题。本文的主要工作包括以下三个方面：一是将多维偏好分析技术与现有软对策模型结合起来．弱化原来要求局中人对所有可行结局排序的条件，使之在分析面向目标的冲突问题中显得更加方便；二是考虑到整体利益与局部利益的差异的普遍存在，且在局中人现有偏好结构下又常无法达到整体最优结局的问题，在冲突分析中引入整体协调人，给出一个二级递阶模型；三是借助个体偏好偏离度向量定量测度个体偏好与整体偏好的差异程度。最后结合一水资源冲突问题进行实证分析。

A Two-Level Drama and Its Applications in Conflict Analysis

Drama theory(DT), suggested by Nigel Howard in 1990, has revised and extended classical metagame in many (aspects) to make it more practical. But it is also based on utility like other game theory and does not deal with the deviations of individual preferences from the general. First, combining Linear Programming Techniques for Multidimensional Analysis with the given model makes DT more convenient to study objective-oriented decision-making because it eliminates the (prerequisite) that asks the players to index all feasible consequences. Then a two-level drama is built up by adding a (coordinator) who represents general interests to get overall optimal resolution due to the ubiquity of deviation of individual (interests) from the general. After that, a preference deviation rector is suggested to measure it. Finally a water resource (conflict) is analyzed to demonstrate its reasonability.