摘要
DSS have been taken as hopfull supports tools for decision making for more than 20 years. There are a lot of literatures on DSS, but unfortunately most of them are not so practical as the designers expected. This paper points out the crux of this situation and argues that the research on DSS should pay some more attention to the decision making activities before the model using stage. A method named "Problem Situation Decomposing Graph (PSDG)" is presented in this paper for helping the decision maker (DM) elicite the problem systems. A PSDG is an acycle AND/OR logical directed graph, which includes all the factors, based on the DM’s knowledge, affecting the problem situation. The logical relationship and utility determining methods in a PSDG can reflect the DM’s decision making style. A method named "Critical Factor Method (CFM)" aided by a logical graph searching algorithm is used to analysis and derive the critical factors which the DM should take into account with priority, and thus supports him to find the decision making goals half artificially and half automatically.
DSS have been taken as hopfull supports tools for decision making for more than 20 years. There are a lot of literatures on DSS, but unfortunately most of them are not so practical as the designers expected. This paper points out the crux of this situation and argues that the research on DSS should pay some more attention to the decision making activities before the model using stage. A method named "Problem Situation Decomposing Graph (PSDG)" is presented in this paper for helping the decision maker (DM) elicite the problem systems. A PSDG is an acycle AND/OR logical directed graph, which includes all the factors, based on the DM’s knowledge, affecting the problem situation. The logical relationship and utility determining methods in a PSDG can reflect the DM’s decision making style. A method named "Critical Factor Method (CFM)" aided by a logical graph searching algorithm is used to analysis and derive the critical factors which the DM should take into account with priority, and thus supports him to find the decision making goals half artificially and half automatically.
