In addition, to provide the most direct comparison
to the results of Figure 1, equal weight (0.25) was
assigned to each criterion with the wildlife reserve
constraint acting as an absolute barrier to development. The result of the averaging process is
shown on the lower left. The image on the lower
right shows the result of selecting the best areas
from this suitability map in order to match the total
area of that selected by Boolean analysis in Figure 1.
Note that as in Figure 1, the distance to labour force
was calculated from a cost distance surface that
accounted for road and off-road frictions.
The continuous suitability map shown in Figure 2
has the same numeric range as the standardised
factors if the weights that are applied sum to 1.0.
A specific decision can then be reached by rank
ordering the alternatives (in this case, pixels) and
selecting as many of the best ranked areas as is
required to meet the objective of the analysis in
question. In Figure 2, this has been done in order to
select as many of the best areas as were selected by
the Boolean analysis in Figure 1.
This procedure of weighted linear combination
dominates multi-criteria approaches with rasterbased
GIS software systems. However, there are a
number of problems with both approaches to multicriteria
evaluation