I have a list of addresses of people participating in carpools to a workplace and a list of cars given by the workplace. Both lists change daily.

I need to group people by their addresses and assign them cars by the group size (and then calculate their route to work, but that's the easy part).

How can I group points (worker addresses) by their geographic location, or better yet, by their closeness on the road network?

I'm using ArcInfo 10.1. I tried using the Grouping Analysis (which groups by geographic location and distance from the workplace), but it's giving me weird results.

closed as unclear what you're asking by PolyGeo Jun 11 '16 at 9:04

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    Can you expand a little bit on the "weird" results -- not what you expected? completely wrong? groups too large? etc. :) – Erica Oct 28 '14 at 15:05
  • 1
    It may be the case that "weird results" are unavoidable, and some level of human oversight is necessary. Eg if there are 6 people in a town, and the only car available fits 5 people, 1 person will be left out - no matter how smart your algorithm – Stephen Lead Oct 30 '14 at 23:28
  • 4
    Only the Network Analyst extension is required to solve these kinds of problems. Use Solve Vehicle Routing Problem for a geoprocessing service and Make Vehicle Routing Problem Layer tool for doing Desktop Analysis. Both create multiple routes to optimize the use of a fleet of vehicles. The extension is not free. – Richard Fairhurst Nov 2 '14 at 0:52
  • 2
    This link explains the Make Vehicle Routing Problem Layer help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//… The problems are described in terms of a commercial fleet, but the restaurant grease pick up problem is similar to what you are trying to do, except with people instead of grease. – Richard Fairhurst Nov 2 '14 at 0:59
  • 1
    The main challenge is how to match between people to share a ride, and to decide who picks up whom with their vehicle, so that it will be worthwhile both for the driver and all the passengers. Even if we assume that every driver can pick up at most one other passenger, the problem is challenging. – Geog Nov 5 '14 at 9:18