Spatial Multi-criteria Evaluation
Why?
There are 12 different inidcators that make up the final TOD score. These 12 maps need to combined into one composite map where each raster cell contains a TOD value. One can then see how the level of TOD varies across the study area.
A SMCE consists of:
-
Building a criteria tree,
-
Standardising each map,
-
Determining map weightings, and
-
Obtaining the final composite map.
Developing different visions
Five different visions were created; each vision varied the weightings of the social indicators while the weightings of the built environment indicators were kept the same.
Varying the weightings of the social indicators allows one to see the effect it has on the final TOD map. Three of the most interesting visions will be covered in this site. These are:
-
Exclusion of social indicators,
-
Emphasis on safety, and
-
Emphasis on car ownership.
This is the list of individual TOD indicators
Three-level hierarchy of the criteria tree.
This is the list of individual TOD indicators
Criteria tree
The 12 indicators were structured into a hierarchy with three levels. The first level groups the indicators into "Built environment" and "Social". The second level organises the built environment indicators into 5 dimensions.
Standardisation functions
Each indicator map has to be standardised so that values are unit-less and range from 0 to 1. This means that the map layers can then be added together.
Weightings
The first vision places a zero weighting on the social indicators - only the built environment indicators are considered. The second vision places emphasis on safety, and the third vision emphasises car ownership. The same standardisation functions and built environment indicator weightings were used throughout.
The pairwise comparison method (Malczewski, 1999) was used to obtain the weightings. The meta-analysis of Cervero & Ewing (2010) was used for the built environment pairwise comparisons.
Obtaining a final map
This slideshow is an example of the process that ILWIS runs once the indicators have been standardised and assigned weightings.
This image illustrates obtaining the aggregated weightings for each individual map layer. The aggregate weighting is the result of multiplying together the weightings at each level of the hierarchy.
This image an example of how the TOD score for one raster cell (i) is calculated by adding the weighted map layers.
Here each raster cell, from i to n, contains a particular value that is the result of adding all the weighted map layers together. Final scores range from 0 to 1.
This image illustrates obtaining the aggregated weightings for each individual map layer. The aggregate weighting is the result of multiplying together the weightings at each level of the hierarchy.
Weighting combination for 'Exclusion of social indicators'
Weighting combination for 'Emphasis on safety'
Weighting combination for 'Emphasis on car ownership'
Weighting combination for 'Exclusion of social indicators'
Click image to enlarge
Click image to enlarge
Click image to enlarge
Click image to enlarge