Clustering locations based on distance

[xuerebry]

I have a simple data set with 3 fields - location name, latitude, and longitude.  I have about 1,100 locations and want to cluster them into regions of about 15-16 locations each.  The furthest distance I want to allow between any 2 locations within a region is 30 miles.  What is my best approach to solving this problem?

[PGStats]

One way is to convert the lat-long to UTM coordinates and use euclidian distance for clustering (the distance between UTM coordinates approximates the local distance in meters).

To do the conversion, look at

https://communities.sas.com/t5/SAS-Communities-Library/Convert-LAT-LONG-to-UTM-and-back/ta-p/221779