ModelBuilder – Troubleshooting ModelBuilder Speed Issues in ArcGIS Desktop

arcgis-desktoparcgis-progeoprocessingmodelbuilder

I've built out a model but it takes what feels like a very long time to run. About 60 seconds for each iteration and I only have 200 records. I am not sure what is normal or expected for these types of things? No tool in particular is drastically slower than the others, it's just all slow. I also feel like just a couple days ago virtually identical model was running much faster. It gets slower every day.

Does anyone have guidance on how to improve processing speed or are there other things in the background aside from the model itself that drastically slow this done? Looking to do anything I reasonably can to speed this up.

I've even tried coding some of it in ArcPy just using the same tools, testing out what might work. It ran even slower.

Attached is an image of the ModelBuilder and here is a list of the tools:

1) Active: Point Feature Class, approx 200 records

2) Iterates through each point

3) Pairwise Intersection to Identify the corresponding Parcel Record

4) Buffer .5 Mile Out

5) Intersect with Roads and School Zone polygon

6) Select which polygon layer the Selected Feature does not exist in (invert selection)

7) Delete the other polygons

8) Clip all other parcels that fall within our remaining polygon

9) Finish

In a nutshell identifying all parcels around a subject parcel that are within the same school zone and do not cross particular highways.

Best Answer

Slowness can be caused by many things and may not even have anything to do with your model, for example your network is slowing down for some reason, or there is not enough space on your hard disk if you are writing locally?

So basically no one will be able to answer this question as no one can observe the performance.

That said there are a few top tips to speed things up:

If any of your input data is a shapefile make sure it has a spatial index, even better move it into a file geodatabase.
If you are using queries to select data then add an attribute index to the field you are querying.
Write intermediate data to IN_MEMORY. It appears you are using ArcGIS Pro, if you have the latest version then use memory instead

Following these top tips you'll at least know that you have fine tuned your model to run as fast as it can and any slowness is probably down to some other system issue or quirkiness with the data.

Related Solutions

ArcGIS Desktop – Finding Parcel Polygon Features with Frontage on Multiple Streets

If your road network is divided by segments, dissolve the buffers based on road name or some common attribute so that individual segments of roads do not count as two separate roads in the next step.
Buffer your street centerlines by a standard amount - more than half the typical ROW width, but no so large that it would overlap parcels that do not front on it. You may want to choose different buffers for different classes of roads. When you buffer, don't merge the polygons.
Spatially join the parcels to the road buffers, choosing "one to many". The "count" field on the results should allow you to identify all parcels with two frontages.

Now you have a table with rows for every overlap between a parcel and a road. You could easily summarize this to find all parcels overlapping two or more roads using either GIS or a database. But eliminating corner lots will be more tricky. Here's one broad method:

Set up topology for the buffer layer with a rule "must not intersect".
Export all errors to a new feature class of intersections.
Spatial join the intersections to the buffer layer, choosing "one to many".

At this point you have two tables: one with all frontages and one with all intersections between roads. It won't be easy, but you can write a SQL query that selects finds all parcels that have two frontages where those frontages do not intersect. The results would be the set of all parcels that have multiple frontages that do not intersect.

One advantage of this method is that it avoids choosing parcels with edges that are not on roads, such as where a parcel is adjacent to an alley, a water features, a railroad ROW, etc. A disadvantage is that the buffering won't be perfect, and you'll still have to visually check it for errors.

I'm not sure how to write the SQL - perhaps someone else can chime in on that.

Edit Thoughts on the SQL:

It would be fairly simple to solve for parcels with two frontages as below. Then you might need to manually look at parcels with more than two frontages, but hopefully there would be few, and most would be corner lots anyway.

This might work for selecting only non-corner parcels with two frontages:

select ParcelID from parcels
where parcelID not in (
    ((select Parcels1.parcelID, intersections.intersectionID from
     intersections left join
        (Select first(RoadID) as Road1, last(RoadID) as Road2, count(*) as frontages, ParcelID
        from Parcels
        where count(*) = 2
        group by ParcelID) as Parcels1
    on intersections.RoadID = Parcels1.Road1) as int1
inner join
    (select Parcels2.parcelID, intersections.intersectionID from
    intersections left join
        (Select first(RoadID) as Road1, last(RoadID) as Road2, count(*) as frontages, ParcelID
        from Parcels
        where count(*) = 2
        group by ParcelID) as Parcels2
    on intersections.RoadID = Parcels2.Road2) as int2
on int1.intersectionID on int2.intersectionID))

ArcGIS for Desktop – Finding Differences Between Shapefiles

With the Compare Feature Tool, you should just choose your Segment_ID field as the sort field in the dialog.

The [sort] field or fields [are] used to sort records in the Input Base Table and the Input Test Table. The records are sorted in ascending order. Sorting by a common field in both the Input Base Features and the Input Test Features ensures that you are comparing the same row from each input dataset.

If you are wanting to get the Segment_ID field as part of your output table, try joining the output Compare Feature table to your source Test Table on cf.Object_ID=tt.Object_ID, then export out the joined table with your Segment_ID.

You could easily put this into a model for easier repeatability.

Best Answer

Related Solutions

ArcGIS Desktop – Finding Parcel Polygon Features with Frontage on Multiple Streets

ArcGIS for Desktop – Finding Differences Between Shapefiles

Related Question