Lab 8: Non-Boolean Multi-Criteria Evaluation

1.

The advantage of WLC is that it gives us the ability to assign different weights to our suitability map factors. Tradeoff weights are assigned to every factor which allows us make operations that are a risk balance between the AND operation and the OR operation. Boolean MCE is either extreme risk aversion or extreme risk taking which makes it not as good for comparing different areas of varying suitability. For example: we are presenting the idea of a new library to some investors. By using WLC we can assign tradeoff weights to the factors that they find most important and a decision of where the best suitably area for the library is agreed upon collectively.

2.

What is meant by this is that continuous factors can be defined by standardized variables, with comparable suitability values on a range from 0.0 – 1.0. By selecting standardization parameters in the FUZZY module, we can apply our knowledge of how suitability changes for different factors to make useful decision for developers. For example: residential development near open waters has a very low suitability in the first 100 meters, then it increases with distance up to 800 meters, and beyond that suitability increases just marginally.

3

There are 3 examples of this type of fuzzy standardization in TerrSet: Sigmoidal, J-shaped and Linear. These ways of standardization can be considered aggregation procedures and they are beneficial because they allow us to retain the variability from our continuous factors.

3.

Re-scaling is a way of changing between quantitative and qualitative data. The FUZZY module is one way to change categorical data into continuous data. When scaling from categorical data to continuous data you are taking a finite number of classes and you are stretching them out to infinity across the entire page. because it takes classes with sharp boundaries and creates a gradual transition of classes ranging form 0.0 to 1.0 Running MCE with WLC creates a multitude of classes ranging across as scale from 0.0 to 1.0. To sum it up: this is an aggregation technique is useful because it turns categorical data into continuous.

4.

In my final project I will be doing a suitability analysis of potentially new areas for a new school in Surrey. In my opinion, using Ordered Weighting Averaging (OWA) could be more appropriate for this. The advantage of OWA is that it allows us to control the level of risk in our MCE to the degree of how much factor weights and our standardized continuous factors can influence the final suitability map. OWA would be beneficial because I would like to use it for comparing the weight of proximity to large residential lots, to medium residential lots, and to small residential lots. This is useful because it will let me because I think schools need to be built in more populated areas to prevent overcrowding in the already built schools. it will let me look at how much the residential building will affect the area of the most populated areas.

5.