Skip To Content

Rank and score sites using suitability analysis

Use Suitability Analysis to rank and score sites based on multiple weighted criteria. Suitability can be ranked based on data variables from Esri’s living Atlas of Demographic and Socioeconomic data and your site attributes. Once you select your criteria, you can assign weights to them, get weighted scores for each potential site, and review their final score ranks from most suitable to least suitable site. Suitability Analysis can be run on a set of point location sites, polygon areas, standard geographies, or any combination thereof.

The following guide will lead you through an example of how to execute this workflow.

From the Create Maps from Data tab click Suitability Analysis.

Launch Suitability Analysis

You will see the Introductory screen that gives you a very brief overview of the Suitability Analysis workflow. If you do not wish to see this screen every time, select Skip this in the future, before you click Get Started.

Get started

For the purposes of this example, you are a real estate agent who has taken on a new client interested in buying a home. Your client is a young professional with an advanced degree, and has an elderly parent living with him who relies on being in close proximity to a senior center for socializing and leisure activities. Your client wishes to buy a home to reside in, that is within walking distance to one of his preferred senior centers, so that his parent has convenient access. You need to present him with a neighborhood analysis for these senior center locations, to help evaluate their suitability based on his criteria:

  • Greater number of individuals with advanced degrees
  • Lower median home value, which would indicate greater affordability of purchasing a home
  • A median age of the population close to his own age (32)

To begin the Suitability Analysis workflow you will select the sites that you want to use in the analysis. For this walkthrough, five sites were already loaded on the map, each representing an area 0.75-mile radius around a senior center.

  1. Click Start with items on the map. Alternatively, we could click Add sites from project, to add sites not already on the map.
    Define your site
  2. The resultant map with your five sites is shown.
    Candidate sites
  3. Click Next.
  4. Expand the Add Criteria drop-down and select Add variables from data browser.
    Add variables
  5. You will select variables using the Data Browser.
    Data Browser
  6. Select the following three variables and click Apply.
    • 2016 Education: Grad/Professional Degree
    • 2016 Median Home Value
    • 2016 Median Age
  7. After selecting the criteria the sites are automatically scored. By default, all three variables are weighted equally and Influence is set to Positive.
    Applied sites
    • You want a site to score higher if it has greater number of individuals holding Grad/Professional degrees, so you will keep the Influence setting Positive for that variable.
    • For Median Home Value, a lower value is more desirable as it is indicative of greater home affordability. Therefore, you will change the Influence setting to Inverse.
    • For the Median Age, the closer the value is to 32 the more ideal the site. Therefore, you set the Influence to Ideal. Use the slider to specify the ideal value of 32.

    You may also adjust the thresholds to exclude sites from the Results table that do not meet must-have criteria for a variable. E.g. you may set a maximum value of $500,000 for Median home value so sites that have a median home value above $500,000 are excluded. Because you are not scoring too many sites in this example, you will not adjust the threshold values and keep the default setting to include their full range.

    The color ramp is applied to color-code the sites on the map by their final score. You may click on the View Results Table link to open the Results Table.

    Results table
    With the above settings, Candidate 2 receives the highest final score in the Results Table.

  8. Assign a Weight of 60% to 2016 Median Home Value. You want to do this as home affordability is a higher priority for your client over his other preferred criteria and it will be a better use of his time to look for a home in the vicinity of those senior centers where home prices are lower.
    Assign variable weights
  9. Assigning weight to one variable will proportionately reduce the weights to the other variables. In this case 2016 Education Grad/Professional Degree and 2016 Median Age have been reduced to 20% each, as the weights must always add up to 100%.
    Using the above weight adjustment - the site, Candidate 3, receives the highest score in the Results Table.
    Note:

    Further details regarding how the below scores are calculated can be found in this section.

    Results table after weight adjustment
  10. To show sites that meet a certain minimum score. In this case, say, only those sites with a score > 0.5. Click the filter icon Filter
    Filter criteria
  11. Drag the slider for the lower limit from 0 to 0.5 (or type the value 0.5 in the text box).
    Filtered results
  12. As we see above, the Results Table is filtered accordingly to show only those sites that have a final score > 0.5. You may click Export… to export the results to an Excel file, or to a new layer that you can also share with others in your organization.

Weighted score calculations

The following are details regarding how the above scores are calculated, using some of the values in the annotated table above to illustrate. Each weighted score is calculated as a percent difference of the value for a given site compared to the target value selected by the user. Here, GP stands for the Grad/Professional degree variable, HV stands for the Median Home Value variable and MA stands for the Median Age variable.

As the above table illustrates, Candidate 3 site has the highest suitability score of 0.86 (Cell 1A). This score is calculated by adding the weighted scores for each of the three variables (GP, HV, MA) used in the analysis.

  1. First, we will examine how the number of households with graduate degrees (GP) contributed to this score.

    • Site Candidate 3 has 1013 households with Grad/Professional Degree (Cell 1B).
    • The maximum value for GP across all the sites is 1272 households for the site Candidate 2 (Cell 2B).
    • Similarly, the smallest value for GP across all the sites is 821 households for site Candidate 1 (Cell 3B).

    In this example, the greater number of people with graduate degrees is desired. This is a positive relationship, so these values are plugged into this formula to calculate the score for GP for Candidate 3:

    Candidate 3 desired values

    We can calculate what the score is for GP for Candidate 3 using the values outlined above:

    Candidate 3 values

    This means, that Candidate 3 has a score of 0.43 (Cell 1C). Once the score is calculated, the weight is then applied to the value to determine how much GP will contribute to the total suitability score for the site. In our example, a weight of 20% was applied to GP. Therefore, the weighted score for GP is calculated as 0.09 (Cell 1D):

    0.20*0.43 = 0.09

    The whole weighted score calculation for Candidate 3 can be expressed as:

    Candidate 3 weighted score
    Note:

    abs is the absolute value function.

  2. Next, let us see how the Median Home Value of the site (HV) contributed to the score:

    • The homes within the area of site Candidate 3 have a Median Home Value of $281,545 (Cell 1E).
    • The maximum value for HV across all the sites is $569,638 for the site Candidate 5 (Cell 5E).
    • Similarly, the smallest value for HV across all the sites is $281,545 for site Candidate 3 (Cell 1E).

    Here, a lower Median Home Value is desired, as an indicator of better affordability. This is an inverse or negative relationship, so these values are plugged into this formula to calculate the score for GP for Candidate 3:

    Candidate 3 home affordability setup

    We can calculate what the score is for HV for Candidate 3 using the values outlined above:

    Applying values

    This means, that Candidate 3 has a score of 1 (Cell 1F). Once the score is calculated, the weight is then applied to the value to determine how much HV will contribute to the total suitability score for the site. In our example, a weight of 60% was applied to HV. Therefore, the weighted score for GP is calculated as 0.60 (Cell 1G):

    0.60*1 = 0.60

    The whole weighted score calculation for Candidate 3 can be expressed as:

    Inverse influence for Median Home Value

  3. Lastly, let us examine how the Median Age of the people living in the area (MA) contributed to the score:

    • The population living in the area of site Candidate 3 has a Median Age of 29.9 (Cell 1H).
    • The maximum value for MA across all the sites is 51.5 for the site Candidate 5 (Cell 5H).
    • The minimum value for MA across all sites is 29.9 for the site Candidate 3 (Cell 1H).

    Here, an ideal value of 32 was selected, as a median age closer to 32 is more desirable. These values are plugged into this formula to calculate the score for MA for Candidate 3:

    Median Age desired values

    This means, that Candidate 3 has a score of 0.89 (Cell 1I). Once the score is calculated, the weight is then applied to the value to determine how much MA will contribute to the total suitability score for the site. In our example, a weight of 20% was applied to MA. Therefore, the weighted score for MA is calculated as 0.18 (Cell 1G):

    0.20*0.89 = 0.18

Candidate 3 final weighted score

Note:

In this example we have not adjusted the threshold for any of the variables. If, for example, the threshold had been set such that the GP value for Candidate 3 did not fall within the specified range, then the weight for that variable would default to 0 and effectively GP wouldn’t be used in the final score calculation - filtering would be applied and that particular suitability score would not be used in the final results table.

Final score is calculated as:

Candidate 3 final score

The final score you see in the table is slightly lower, at 0.86, but that's only because it adds the unrounded values instead of the rounded values we used here:

Final score table

This walkthrough gave you a basic understanding of how suitability analysis works. You were able to create a ranked list of the top 3 most suitable senior centers in the area, for your client's weighted criteria from the point of view of house hunting.

You may add additional sites and variables, and further adjust the settings to perform an even more sophisticated suitability analyses. You could further enhance your analysis by including relevant variables that are attributes of the sites being scored. For example, every senior center may have a rating, which is an indicator of quality. You may want to factor that into your suitability analysis, with the Add point layer (e.g. competitor layer) option under Add Criteria.

Add point layer


In this topic