ArcGIS 3.0.2: How does the “service area” behave in zones with no road network?

[JamalNUMAN]

ArcGIS 3.0.2: How does the “service area” behave in zones with no road network?

For example, in the screenshots below, the cost is set to be the distance (road length). I couldn’t figure out how the result of the service area can be explained as the cutoff is set to be 5000, 7000, and 12000. How zones closer to facilities are not covered with the service area while roads that are farther are covered as marked in the screenshots?

[AndreasHall]

I am guessing a bit now but it seems like only blocks that are completely circumfixed by the service area are colored as belonging to that service area. If that is the case your result makes sense. Try adjusting these settings and see if you get a different output that is more in your liking:

[JamalNUMAN]

The secret is still unrevealed.

In the screenshot below, the two issues below are still unexplained:

• How come the zones closed to the facilities are classified as 12,000 meters away from them?
• The two zones marked with angry faces are completely circumfixed by the service area, nevertheless, the zone is not colored!

The data is attached

[AndreasHall]

• The colored zones are not classified as being 12 000 meters away from the facilities, they are classified as belonging to the service area that covers everything that can be reached within 12 000 meters.
• I would guess the zones with angry fazes are circumfixed by two different service areas and that no single 12 000 meter service area is completely covering the area. (I tried opening your data but got an error message regarding your network dataset.)

[JamalNUMAN]

Thank you for the prompt answer.

The result is still unjustified. How come the zones of left angry faces are colored while the ones on the right are not! I have exaggerated the trim distance in order to better present the result.

How this can be explained?

By the way, the network dataset has only the default impedance (length), thus, it can be built easily if it fails to open from the attachment.

[AndreasHall]

I do not have any better answers than the ones I have provided, sorry. I am not on the product team and I do not have insight into the algorithms these tools are using.

[JamalNUMAN]

The “split” option generates different result as appears in the screenshot below. However, either way, I was expecting to have a result like the one I sketched in orange.

The second screenshot shows how the result appears in other roads network. As expected!

[AndreasHall · Accepted solution]

The data you sent me is covering a very large area. The blocks are about 5-8 km wide and 2-5 km high. This might be affecting your results. The other example you refer to seems to be a much denser road network.

[JamalNUMAN]

Thank you AndreasHall for the input.

It does make sense. There should be a relation between the distance between roads and “trim distance”