Fedrizzi and Mich [Fed91] presented a new Group Decision Support System (GDSS) logic architecture in which linguistic variables and fuzzy production rules were used for reaching consensus. In [Fed92] it was shown that when all the fuzzy numbers representing the performance levels have continuous membership function, then the consensus degrees (defined by a certain similarity measure) relative to each alternative are stable under small changes of the experts’ performance levels. Generalizing the method of [Fed91], we represent the knowledge via fuzzy production rules of more complex form, which makes it possible to determine the actual (and overall) group performance level in one step by using Zadeh’s compositional rule of inference in the consensus management module.
Fuzzy Reasoning Techniques for GDSS
Mich, Luisa
1993-01-01
Abstract
Fedrizzi and Mich [Fed91] presented a new Group Decision Support System (GDSS) logic architecture in which linguistic variables and fuzzy production rules were used for reaching consensus. In [Fed92] it was shown that when all the fuzzy numbers representing the performance levels have continuous membership function, then the consensus degrees (defined by a certain similarity measure) relative to each alternative are stable under small changes of the experts’ performance levels. Generalizing the method of [Fed91], we represent the knowledge via fuzzy production rules of more complex form, which makes it possible to determine the actual (and overall) group performance level in one step by using Zadeh’s compositional rule of inference in the consensus management module.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione
