Multi-criteria evaluation in GIS is concerned with the allocation of land to suit a specific
objective on the basis of a variety of attributes that the selected areas should possess.
Although commonly undertaken in GIS, it is shown that the approaches commonly used in
vector and raster systems typically lead to different solutions. In addition, there are
ambiguities in the manner in which criteria should be standardised and aggregated to yield
a final decision for the land allocation process. These problems are reviewed and the
theoretical structure of fuzzy measures is offered as an approach to the reconciliation and
extension of the procedures currently in use. Specifically, by considering criteria as
expressions of membership in fuzzy sets (a specific instance of fuzzy measures) the
weighted linear combination aggregation process common to raster systems is seen to lie
along a continuum of operators mid-way between the hard intersection and union operators
typically associated with Boolean overlay in vector systems. A procedure for implementing
this continuum is reviewed, along with its implications for varying the degrees of
‘ANDORness’ and trade-off between criteria. In addition, the theoretical structure of fuzzy
measures provides a strong logic for the standardisation of criteria and the evaluation of
decision risk (the likelihood that the decision made will be incorrect)