有限理性模型查看源代码讨论查看历史

         

有限理性模型

[1]来自 搜狗 的图片

有限理性模型是指20世纪50年代之后，人们认识到建立在“经济人”假说之上的完全理性决策理论只是一种理想模式，不可能指导实际中的决策。赫伯特·西蒙（Herbert Simon）提出了满意标准和有限理性标准(Bounded Rationality Model)，用“社会人”取代“经济人”，大大拓展了决策理论的研究领域，产生了新的理论——有限理性决策理论。 有限理性模型又称西蒙模型或西蒙最满意模型(Simmon's Bounded Rationality Model)。这是一个比较现实的模型，它认为人的理性是处于完全理性和完全非理性之间的一种有限理性。

主要观点

（1）手段-目标链的内涵有一定矛盾，简单的手段-目标链分析会导致不准确的结论。

西蒙认为，手段-目标链的次序系统很少是一个系统的、全面联系的链，组织活动和基本目的之间的联系常常是模糊不清的，这些基本目的也是个不完全系统，这些基本目的内部和达到这些目的所选择的各种手段内部，也存在着冲突和矛盾。

（2）决策者追求理性，但又不是最大限度地追求理性，他只要求有限理性。

这是因为人的知识有限，决策者既不可能掌握全部信息，也无法认识决策的详尽规律。比如说，人的计算能力有限，即使借助计算机，也没有办法处理数量巨大的变量方程组；人的想像力和设计能力有限，不可能把所有备择方案全部列出；人的价值取向并非一成不变，目的时常改变；人的目的往往是多元的，而且互相抵触，没有统一的标准。因此，作为决策者的个体，其有限理性限制他作出完全理性的决策，他只能尽力追求在他的能力范围内的有限理性。

（3）决策者在决策中追求“满意”标准，而非最优标准。

在决策过程中，决策者定下一个最基本的要求，然后考察现有的备择方案。如果有一个备择方案能较好地满足定下的最基本的要求，决策者就实现了满意标准，他就不愿意再去研究或寻找更好的备择方案了。

这是因为一方面，人们往往不愿发挥继续研究的积极性，仅满足于已有的备择方案；

另一方面，由于种种条件的约束，决策者本身也缺乏这方面的能力。在现实生活中，往往可以得到较满意的方案，而非最优的方案。

根据以上几点，决策者承认自己感觉到的世界只是纷繁复杂的真实世界的极端简化，他们满意的标准不是最大值，所以不必去确定所有可能的备择方案，由于感到真实世界是无法把握的，他们往往满足于用简单的方法，凭经验、习惯和惯例去办事。因此，导致的决策结果也各有不同。

原文

Herbert Simon (1916-2001) is most famous for what is known to economists as the theory of bounded rationality, a theory about economic decision-making that Simon himself preferred to call “satisficing”, a combination of two words: “satisfy” and “suffice”. Contrary to the tenets of classical economics, Simon maintained that individuals do not seek to maximise their benefit from a particular course of action (since they cannot assimilate and digest all the information that would be needed to do such a thing). Not only can they not get access to all the information required, but even if they could, their minds would be unable to process it properly. The human mind necessarily restricts itself. It is, as Simon put it, bounded by “cognitive limits”.

Hence people, in many different situations, seek something that is “good enough”, something that is satisfactory. Humans, for example, when in shopping mode, aspire to something that they find acceptable, although that may not necessarily be optimal. They look through things in sequence and when they come across an item that meets their aspiration level they go for it. This real-world behaviour is what Simon called satisficing.

He applied the idea to organisations as well as to individuals. Managers do much the same thing as shoppers in a mall. “Whereas economic man maximises, selects the best alternative from among all those available to him,” he wrote, “his cousin, administrative man, satisfices, looks for a course of action that is satisfactory or ‘good enough’.” He went on to say: “Because he treats the world as rather empty and ignores the interrelatedness of all things (so stupefying to thought and action), administrative man can make decisions with relatively simple rules of thumb that do not make impossible demands upon his capacity for thought.” The principle of satisficing can also be applied to events such as filling in questionnaires. Respondents often choose satisfactory answers rather than searching for an optimum answer. Satisficing of this kind can dramatically distort the traditional statistical methods of market research.

Simon, born and raised in Milwaukee, studied economics at the University of Chicago. “My career,” he said, “was settled at least as much by drift as by choice”, an undergraduate field study developing what became his main field of interest—decision-making within organisations. In 1949 he moved to Pittsburgh to help set up a new graduate school of industrial administration at the Carnegie Institute of Technology. He said that his work had two guiding principles: one was the “hardening of the social sciences”; and the other was to bring about closer co-operation between natural sciences and social sciences. Simon was a man of wide interests. He played the piano well—his mother was an accomplished pianist—and he was also a keen mountain climber. At one time he even taught an undergraduate course on the French Revolution. He was awarded the Nobel Prize for economics in 1978, to considerable surprise, since by then he had not taught economics for two decades.

有限理性模型的两种决策方法

一、组合的排列法

从工作任务的不同组合排列中进行选择，以达到满意结果的一般性决策方法。

二、方面排除法

这是一种适用于对决策方案进行同时性选择的有效方法。