Gravity Models: Predicting Attraction by Population Size


This post may describe functionality for an old version of CARTO. Find out about the latest and cloud-native version here.
Gravity Models: Predicting Attraction by Population Size

In our previous blog post we introduced Gravity Models a useful function that allows businesses and organizations to choose the location of their next store or mall. This time we are going to apply Huff’s Law to populated areas to study their attraction based on population as an extension to the gravity model of migration.

According to this law the probability of patronage to a specific business is dependent on the distance from the patron’s home to that business the attractiveness of that business based on the size of the surrounding population and adjusted by the presence of other potential businesses in the neighborhood.

When tasked with selecting a new location for your next store or business you can use these models as a way to predict potential success for profit and popularity. For example in this visualization by selecting Zaragoza from the drop down list of cities in the top right selector and zooming in you'll notice several factors:

  • Placement: Managers typically place their businesses in locations with more than 20 000 inhabitants. In the case of Zaragoza there is a potential customer count of ~1 million. If these results align with the project's requirements you can adjust attributes and make an even more strategic decision based on the analysis disccussed in the previous post.
  • Traffic forecast: If you narrow the probability range to above 90% the tool renders the populated places where people are willing to commute to Zaragoza to shop. By checking the top places you can decide where to add new bus lines.
  • Understanding the demand for public services requires forecast analysis: In order to scale up the current public service provision in Zaragoza you can have an initial estimate of the potential number of citizens that will use these services in upcoming years: ~1 million vs. the current 680 000 inhabitants.

Gravity Models make decision making processes easy when location is key to your businesses success. Stay tuned for our next installment on Gravity Models and how you can optimize your strategic location-based business decisions.

Happy data mapping!