A neural network technique in modeling multiple criteria multiple person decision making

Neural networks which use the back-propagation learning algorithm under monotonic function constraints can be used in modeling multiple criteria multiple person decision making (MCMPDM). This is done by training the neural networks with the judgment data of a set of individual decision makers, thus aggregating and generalizing their decision making knowledge. The generation of monotonic value functions in MDMPDM is demonstrated, and the representation of uncertainty using the fuzzy characteristics of MCMPDM is also illustrated.