I'm working with a set of 88 counties in Ohio, and I need to cluster them into groups of 3 based on minimum aggregate distance between each possible cluster of 3 (I created polygon centroids for each county). This will be done without replacement. Once a county is assigned to one cluster, it is not available for others. Put another way, if there is A, B, C, and D in close proximity to one another, does A/B/C or B/C/D create "tightest" group based off sum of the distances between A-B, B-C, and A-C versus B-C, C-D, and B-D. I am aware of the Grouping Analysis tool in version 10.1 (not positive that would do the trick...), but unfortunately I do not currently have access to that platform. Help please! Thanks in advance.
... View more