The most common approach is to
rescale the range to a common numerical basis by
simple linear transformation (Voogd 1983), as was
applied in Figure 2. However, the rationale for doing
so is unclear. Indeed, there are many instances where
it would seem logical to rescale values within a more
limited range. Furthermore, there are cases where a
non-linear scaling may seem appropriate. Since the
recast criteria really express suitability, there are
many cases where it would seem appropriate that
criterion scores asymptotically approach the
maximum or minimum suitability level.
The third issue concerns the weights that are
applied. Clearly they can have a strong effect on the
outcome produced. However, not much attention
has been focused in GIS on how they should be
developed. Commonly they represent the subjective
(but no less valid) opinions of one or more experts
or local informants. How can consistency and overt
validity be established for these weights?
Furthermore, how should they be applied in the
context of varying trade-off between factors?
A fourth problem concerns decision risk.