We have a complex problem to solve using our health care data. We want to take each patient and calculate driving distance (miles) to each of our 40 locations. I have the lat/long for all patients and all locations. Also, I have the network street data. I've never used network analyst so I am not sure it can handle this. Does anyone have any experience doing this type of routing? Any thoughts would be greatly appreciated.
Do you want to assign patients to each of the 40 sites based on the shortest distance between the sites? Like a service area? Service area analysis—Help | ArcGIS Desktop or do the patients have to go to particular sites regardless of where they are located?
Thanks for the reply. For this project, we actually want to store the driving distance to each location in our geodatabase. At this point we are not concerned about which location they should go to or which one is closer, we just want to store all of these data points for further analysis.
I'm able to simply this task a bit. What we have is the patient X/Y and location X/Y for every visit. We want to see how far they drove for each visit. We have all the X/Y data and I have the network data, my only question is how do I package this all together to the distance (in miles) for each visit? Any ideas?
Or a closest facility type of problem: you'd need to iterate through all you patients and then route to each of the 40 locations; that'll give you the travel distance by patient. So long as you have your network set up to calculate travel distance.
Some links:
I'd probably go with Route or Closest Facility. See the examples at:
ArcGIS Help (10.2, 10.2.1, and 10.2.2) - Types of Network Analysis
One could do the iteration though each record with Modelbuilder or Python.
More helpful links:
ArcGIS Help (10.2, 10.2.1, and 10.2.2) - What is Network Analyst
Creating a network dataset—Help | ArcGIS for Desktop
Caution:
Keep in mind that this analysis requires really good data and that many datasets (like street centerlines) that seem like they would be easy to use actually require considerable cleanup before the analysis can be run. Minute gaps, like those found in many CAD-derived street centerlines - will stop an analysis cold. Don't be surprised if you find that dozens of hours of cleanup are required before one can push the Solve button and a minute later get the results. Also, the Network Analyst extension is not the easiest extension to learn - it has many oddities.
Chris Donohue, GISP
as for the OD Matrix since you are only interested in the result and not the route... note this comment
When you choose to display lines in the OD cost matrix analysis layer, they appear as straight lines. However, their cost attributes always report the least-cost network path. The following table describe the output fields of the Lines feature class:
Caution:
Keep in mind that this analysis requires really good data and that many datasets (like street centerlines) that seem like they would be easy to use actually require considerable cleanup before the analysis can be run. Minute gaps, like those found in many CAD-derived street centerlines - will stop an analysis cold. Don't be surprised if you find that dozens of hours of cleanup are required before one can push the Solve button and a minute later get the results. Also, the Network Analyst extension is not the easiest extension to learn - it has many oddities.
^^^^^^ This... ^^^^^^^