The use of multi-criteria evaluation in spatial policy

Conventional multi-criteria evaluation methods have largely been non-spatial. They use average or total impacts that are considered appropriate for the entire area under consideration. However, the implicit assumption that the study area is spatial homogenous is not always realistic because in many cases the effects vary across space. Space is an important dimension in many decision problems, therefore this paper shows how multi-criteria evaluation methods can incorporate spatial effects and to support spatial policy. The paper starts with the definition of a number of basic concepts such as spatial criteria, spatial decision problems and spatial multi-criteria methods. The main purpose of these methods is to structure and aggregate scores reflecting the performance of alternatives on multiple objectives measured across multiple spatial units. The obvious way to carry out such an aggregation is to do it in two steps: first aggregate across criteria, followed by an aggregation across spatial units. But the reverse order is also possible. The first approach is often used in a GIS context by means of overlay techniques. The reverse order approach is usually followed in the mainstream of multi-criteria methods. The present paper discusses the pros and the cons of both approaches.