Research On Method For Multiple Attribute Risk Decision Making With Incomplete Probability Information

In a market economy and social information explosion today, whether it is in the organization system reform, or in market development in many aspects of the emergency plan selection, risk investment, risk decisions are everywhere. Risk decision-making process of selection and decision-making consequences of uncertainty, there are certain risks, but also for risk decision-making. Due to the complexity, volatility of objective things, and limited human experience knowledge, in the actual risk decision making problems, decision makers need to weigh the interests of all parties to take into account various effects, and consider a variety of indicators to achieve the expected outcomes. At the same time, each an indicator may have a different state, and the probability of these states is generally difficult to obtain.Therefore, how to use existing information on the condition that only part of the probability information is given, to make a scientific and effective decision-making method is a meaningful and important research topic.In this paper, on the basis of a comprehensive review of relative methods for risk decision making problems at home and abroad, two main questions on multi-attribute risk decision-making problems with incomplete probability information are analyzed in detail: firstly, how to calculate the probability based on existing information. Secondly, how to choose the optimal alternative on the condition that decision makers have requirements and expectations of attributes.On the basics of existing research, this article studies multi-attribute risk decision making problems with incomplete probability information and the main works are summarized as follows:(1) A kind of the multi-attribute risk decision-making problems with incomplete probability information is refined.According to the risk decision making problems’ complexity and uncertainty factors, the various concepts involved in the research questions, variables and other are defined, and an optimization model of each attribute is built.(2) An algorithm to find Attribute’s state probability is given. Under incomplete probability information in the form given, an optimization model based on prospect theory is built. According to the Principle of comprehensive prospect value the bigger the better, the optimal coordination of the probability vector is solved.(3) A method for multi-attribute risk decision-making with incomplete probability information is proposed. With regard to multi-attribute risk decision-making problems with incomplete probability information, a decision making method based on prospect theory is proposed. This method firstly describes multi-attribute risk decision-making problems with incomplete probability information based on prospect theory. The prospect value and state for every interval value given is calculated, and the order of the program is determined.(4) This article proves that the method for multi-attribute risk decision-making with incomplete probability information in reality is confirmed.Ammonia leak accidents are taken as a practical example to illustrate the feasibility and validity of the proposed methods.Finally, conclusions, research fruits and major contribution of this dissertation are given. On the basis, some suggestions on further work are discussed.