Grouping points on a map based on proximity (distance)

[maxmd]

Hello Alteryx Community!

I have been struggling with a work flow to group points on the map based on how close to each others.

I have data set of many points on the map with longitude and latitude. My goal is to test if any of the points are located within 80 meters (0.08 KM) from each other to group them in one group and find how many points are within that group and label each group appropriately. I created centroid and used trade area of o.08 km around each centroid. My objective is to capture the intersected centroids and label each point in the intersection area with the name of the point and how many points were in the area within  the 80 meter rule.

Can somebody help me please?

 

[Thableaus]

Hi @maxmd 

I'm not so sure how your data looks like, but I think Spatial Match Tool could help you out with this.

Basically the Trade Area would be your Target, containing your Universe (made of Centroid Points).

That means that for each record's Trade Area, if it finds a matched point, it will generate a new record. 

That said, you could group them by an ID or something else.

Try this solution and let me know if you struggle with that.

Cheers,

[PhilipL]

Hi maxmd@mst.edu

A bit late, but also posting for others that may have a similar question.

You could consider grouping the points based on the concentration of points within your 80 meter rings using Spatial Match, and an iterative macro.

It's possible to determine which 80 meter zone has the greatest concentration of points contained within by using a Spatial Match, Summarize (count), Sort (descending) and Sample (1st row). Using an iterative macro, you can then remove the selected points from heaviest concentrated 80m ring from the process and cycle the remaining points back in for additional passes with the same approach.

Repeat this process over and over with multiple iterations until you have assigned all of your points to unique groups.

Hope this helps.

Phil

[Alder]

Hello I managed to do this myself, I think it does exactly what you need it to do.
Even if the same point(s) is in multiple sets it will show up as in more than 1 set

[Alder]

more geo here, without grouping but parsed