摘要：the non-definite multi-attribute decision making problem is the further expansion of research to the traditional multi-attribute decision making problem. In the actual decision-making, because of policy-making question's complexity, decision-making information acquisition costs and other reasons, the decision information which the policy-maker can obtain are mostly ambiguous, so in the real life, many of decision-making problems are the non-definite multi-attribute decision making problems. In the non-definite multi-attribute decision-making process, as decision-making information is non-definite, therefore the policy-maker, when carries on the decision-making, faced with the following difficult position: decision-makers making decision with the non-definite decision-making information most likely will regret. This article will conduct the research to this question, in order to portray the decision-makers dilemma in the decision-making process under the uncertain condition; this paper defines the following concept: policy-makers plus regret degree, policy-makers negative regret degree and policy-makers weighted combination regret degree. According to the different thresholds of policy-makers plus regret degree, policy-makers negative regret degree and policy-makers weighted combination regret degree, respectively determine the best plan and construct the optimal decision results table, policy-makers according to their preference situation can query this table and select the most excellent plan suited to their own situation.
