MapInfo Pro

MapInfo Monday: Proportional Overlay Analysis, Quick & Dirty

  • 1.  MapInfo Monday: Proportional Overlay Analysis, Quick & Dirty

    Posted 05-17-2021 15:57
    Edited by Peter Horsbøll Møller 05-17-2021 15:58

    Happy #MapInfoMonday!

    Today we will inspect one way of doing a proportional overlay analysis. As you probably can guess, there​ are multiple ways of doing this in MapInfo Pro. Here we will look at a quick and dirty way of doing it using some of the built-in commands.

    What is a Proportional Overlay Analysis?

    OK, let's cover the basics first.

    When doing a Proportional Overlay Analysis, you are looking to understand for example the catchment area of a shop. This is often done using a drivetime region where you estimate that the people visiting your shop only want to drive a given distance. You can then create a drivetime region representing this distance which gives you an estimated "catchment zone" for your store.

    The next step is to understand how big a potential customer base this will give you. To do so, you will need to have some demographic data covering your area of interest, typically for the country you are working in.

    The idea is now to analyze the demographic polygons that fall within your drivetime region and calculate how big a percentage of these demographic regions are within your drivetime region. Using this percentage, you can then estimate the number of customers living inside the drivetime region.

    Proportional Overlay Analysis using Erase Outside

    In this example, we are using the Erase Outside command to erase the objects and the part of objects that are outside our "area of interest".

    Below you can see my basic example. I have a circular polygon shown with a red and white line style. That's my area of interest, my catchment area. This polygon can of course take any form or shape. For demonstration purposes, I have kept it very simple.

    I also have a table with demographic information. It's called "Population 2008-2019". 


    1. Make the layer with the demographic data editable via the Layer List in the Explorer window. Then select all objects from the layer. This can easily be done by right-clicking on the layer and using Select All.

    2. Now use Set Target from the Spatial tab to set all the selected regions as your target objects for the next processing. Notice how the target objects have been visualized with a different color on the map.


    3. Select the area of interest. In this case the circular polygon in the map.

    4. From the Erase dropdown on the Spatial tab, select Erase Outside Target. This will delete all records that are entirely outside of the currently selected polygon, and erase the portion of the objects that are partly outside the currently selected polygon.

    Basically, we will end up with only the portion of the objects that are inside the catchment area. This is also what we are looking for as that would be the parts that we would assume would be "visiting" our store.


    5. As part of the Erase Outside process, MapInfo Pro will ask you how you want to handle the attribute information. This is handled via the Data Disaggregation dialog. Here you specify how to handle the values for each column in your table.

    You can choose to keep the value that is recommended for values like names, zip codes, and so on.

    You could also decide to blank out the values of a column. This would be used for columns where you wouldn't need the data anymore, or where the values wouldn't make sense after you have erased part of the object.

    And Finally, you can choose to let the value depend on the Area Proportion of the remaining object. This should be used where the value relates highly to the size of the polygon. Therefore this is useful when calculating the estimated inhabitants in the remainder of the object. Area Proportion is only supported for numeric columns.


    And here is the result of the Erase Outside process.


    Now it is time for a warning. You may not be aware but we have now basically deleted most of the records from your input table, the table with your demographic data. And you would probably not want to lose all this data. But the good news is that we haven't saved these changes to the table, yet.

    So to avoid losing the data, you can do two things:

    1. Now save a copy of your table with the demographic data, and then revert all the changes to the original table.

    2. Or maybe even better save a copy of the table with the demographic data before the processing and use that copy for the processing.

    In the image below, you can see the original table shown below the new table where only the objects inside my area of interest are kept. Notice how the colors of the new table aren't as dark as the colors of the surrounding polygons. That's because the values of these smaller polygons also are smaller.


    I have used the Info tool to query the attributes of one of those polygons. In the Info window, you can see that the original polygon was 26,181. In the polygon inside my area of interest, the value in the same column is now only 19,831 because the size of the remaining polygon is about 30% smaller than the original polygon.


    As I said at the beginning of this post, this is a quick and dirty way of doing proportional overlay analysis. It does however give you the result you are looking for.

    In an upcoming post, I'll show you have you can do a similar process using the SQL language of MapInfo Pro. Stay tuned for this!



    ------------------------------
    Peter Horsbøll Møller
    Principal Presales Consultant | Distinguished Engineer
    Precisely | Trust in Data
    ------------------------------