The analysis: the fun part!  My analysis, as mentioned earlier, consists mainly of creating a series of boolean maps and producing two maps using Multi-Criterion Evaluation.  The final maps that I will produce from this will be maps depicting areas where it is suitable to place the store, one taking the Skytrain into consideration and one that ignores it.  Running the MCE is the easy part: creating the various boolean maps requires a series of data manipulation, value reclassing and assigning, buffer or distance operations if required, and final layer creation using reclass or assign.

       The Major Roads Factor

The first map that I created was a map showing distances from major roads.  I decided to make a map showing distances in equal interval classes of 50 metres, and then a boolean map to use in the MCE with areas less than 250 metres from a road classed as 1 and all other areas classed as 0.  I then added the city vector layer so that at least some figure-ground relationship would exist.  Its hard to tell where the water is on this map, but the main purpose of showing distances from roads is achieved.

To see the roads boolean map I used for analysis, click here
 
 
 
 
 
 
 
 

The next map that I produced is a boolean image resulting from an overlay of two boolean images.  Using the landuse raster coverage, I first created a boolean map showing commercial zones. Then I used ASSIGN to create a map showing medium and high density residential zones.  I then buffered these residental zones, setting the buffer zone and target to 1 and the outlying areas to 0.  This map is a result from multiplying the first map by the second.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

When it comes to saving money, everyone is in on it.  Although generally you would want to set up a store in areas of higher incomes, I have decided to make one of the criteria in my analysis that the area's average income be $60,000 or less.  People with less income SHOULD be more interested in saving money than people with a high income; therefore, they would visit our discount retail outlet and hopefully spend some money!
        For this analysis, I needed some income figures for the GVRD.  I found this in tabular format at their website and adding it to the CITY raster layer produced a rather nice choropleth map.  Although I was only interested in areas of medium and lower income, I decided to show entire map as well to give you an idea of the area.  To see the boolean image, click here.
 
 
 
 
 

        Population Density

This map was also produced from the landuse map in a similar way to the income map.  I got the information from the same website, and used ASSIGN to assign individual population values to each enumeration area.  Then I used AREA to create a map from LANDUSE showing the area within each of these areas.  OVERLAY produced this choropleth, by dividing the population map by the area map, shown here to the right.  To take population density into consideration in the MCE, I created a boolean image that had areas with a density of 2000 people / square km or more to 1 and areas with less to 0.  To see the boolean image click here
 
 
 
 
 
 
 
 
 
 

        The Wal-Mart Factor

A major factor in placing this store is the proximity to other discount retailers.  Although I could probably run an analysis on all the major chain discount stores, I'd probably end up with them covering the entire map.  To prevent the entire GVRD from being declared unsuitable, I've decided only to analyze Wal-Mart locations.  In terms of spatial relationship, I would either want to locate my store right beside a Wal-Mart so we can mooch off of their customer base, or as far away as possible so they don't draw our customers away from our store with their name, Wal-Mart.
I've decided to find areas that are NOT influenced by Wal-Mart.  I incorporated this factor into the MCE by reclassing all areas within 5000 metres as 0 for strong influence (not wanted), and all areas further as 1 for weak influence. To see the boolean image, click here.
 
 
 
 
 
 
 

For the last criterion in my project, I decided to make it an optional criterion.  There really wasn't much to creating this image: I Converted the vector layer SKYTRAIN into a raster layer, then I created SKYBUF, a raster image depicting a buffer of 250 metres around the skytrain line.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

        The Multi-Criterion Evaluation

This part of the analysis involves mashing all of the above created boolean images together and producing the final maps using the MCE function. The first map produced shows all areas that are suitable with respect to the skytrain.  The second map greatly increases the possibilites for a location, for I have removed the skytrain factor.