FUZZY MEASURES AND STANDARDISATION
Clearly, this consideration of fuzzy measures has
implications beyond those of the aggregation
process alone. It also provides a very strong logic for
the process of standardisation. In this context, the
process of standardising a criterion can be seen as
one of recasting values into a statement of set
membership – the degree of membership in the final
decision set. Indeed, Eastman and Jiang (1996)
argue that such statements of set membership in fact
constitute fuzzy sets (a particular form of fuzzy
measure), while those of Boolean constraints
represent classical sets. This clearly opens the way
for a broader family of set membership functions
than that of linear rescaling alone. For example, the
commonly used sigmoidal (s-shaped) function of
fuzzy sets provides a simple logic for cases where a
function is required that is asymptotic to 0 and 1. It
also suggests that the minimum and maximum raw
factor values should not blindly be used as the
anchor points for such a function. Rather, anchor
points that are consistent with the logic of set
membership are clearly superior. For example, in
Figure 4, sigmoidal membership functions were
created for each factor, with anchor points set at the
points where the factor begins to have an effect and
where the effect is no longer relevant. The distance