K-Means Controlling Cluster Size

[CBags]

I'm running a k-means analysis to build out sales territories based on geographic proximity (i.e. stores closest to each other should be in the same cluster), but I'd like to add a limitation on the size of the cluster. For example, I'd like clusters to not exceed 600 stores. Is something like that possible?

[gawa]

hi @CBags 

You cannot specify cluster size because the purpose of K-means cluster analysis is to determine the group based on the given number of clusters so cluster size shall be freedom.

If you already have geometric data(longitude, latitude), you can utilize Spatial Analysis. You can calculate and manage geometric data easily with Spatial tools. Please consider this option, too.

[CBags]

Thanks gawa. I'm not super familiar with spatial tools. Which tool would specifically give me the ability to bucket stores based on proximity (including other limiting factors like total group size)?