Latest Contributions by MelindaMorang

For your first question, if you have time, would you be able to share your data and workflow steps with me as is so I can try to track down that Error 999999?  We are always trying to flush those out and get rid of them, so if you have a consistently reproducible instance of it, I'd like to log a bug for us to fix it.  I can't tell from your description so far whether this happened when creating the network dataset, solving, adding locations, etc. But, ultimately you need a way to get up and running with your Service Area analysis.  If you haven't already, please walk through the tutorial for creating network datasets.  You might also find useful this set of downloadable Tasks for creating a local government network dataset from street centerlines. If those resources don't fix the problem, please post more detail about exactly where the problem is occurring and share your data if you can.  (You can message me privately if you don't want to share your data publicly.) For your second question, you can automate Service Area calculations using Model Builder or Python.  Here are some resources that will help: Model Builder tutorial General how-to for network analysis with Python Python script tool tutorial ... View more

Yes, the Network Analyst Service Area solver seems like the correct approach. Here's a tutorial to get you started: https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/service-area-tutorial.htm  Note that Service Area gives you the option to produce two different outputs: polygons and lines.  The polygon output visualizes the area that's reachable, similar to a buffer but representing travel constrained to the road network.  The line output returns the actual road (or sidewalk) segments in your network that were reached.  You may want to experiment with both to see what works best for your needs. Key for you will be creating (or purchasing) a network dataset that adequately represents pedestrian walking paths, particularly for an analysis as fine-grained as 1000 meters.  I'm not sure our Streetmap Premium data will be adequate.  It contains some information about whether pedestrians can travel on streets and includes some separated walking paths, but it doesn't include sidewalks and crosswalks and stuff like that.  You are likely to have to create a network dataset for your city using sidewalk and street data from the city or county government, if it's adequate.  Here's a tutorial to get you started on how to create a network dataset in general: https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/how-to-create-a-usable-network-dataset.htm You may get some benefit from this set of Tasks for creating a local government network, although the network that comes out of this is fairly simplistic and doesn't include sidewalks: https://www.arcgis.com/home/item.html?id=9e321cf94d2b43b991cc5109189b2c26  ... View more

Unfortunately, no, the situation has not changed.  Supporting passenger-facing routing applications really  just isn't in our road map right now.  If you want to add more detail about what you're trying to do and why other commercially available routing apps aren't adequate for your needs, I can add this info to our collection of use cases, but it seems unlikely that we'll be pursuing this work in the near future. ... View more

Okay.  If you can make that work, then feel free to go for it! However, some data model manipulation might be easier/quicker.  You can fairly easily join the GRouteType field from the Lines table to LineVariants to LineVariantElements. Then just delete all records from those three tables where GRouteType is the types you don't want.  I think then you should be able to run Calculate Transit Service Frequency out of the box and not worry about it picking up the data you want to exclude. ... View more

Thanks for this idea. Off the top of my head, I can't think why this would not be technically possible in a straightforward fashion. The main challenge is in making the tool UI performant. Your specific use case relies on a non-standard route_type value, 712, that is not part of the GTFS specification or Esri's data model.  If you want this to work, the UI would have to open the Lines table and read it to create the choicelist of modes to include/exclude (slow) rather than hard-coding a list of supported mode types (fast).  Not impossible, just not preferred. Another, related enhancement would be to exclude trips and stops that are not accessible to bikes or wheelchairs. ... View more

The Generate Origin Destination Cost Matrix Ready-To-Use tool always uses the ArcGIS Online service, and you cannot change it to use local data.  If you are indeed using this tool, that's why you're still running into the limit. This tool has been deprecated in the last couple of releases of ArcGIS Pro, and confusion over the network data source was one of the reasons.  If you want to do a network analysis, particularly using local data, it's better to use the standard workflow of creating an OD Cost Matrix layer, adding locations, and solving, as described in this tutorial: https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/od-cost-matrix-tutorial.htm (And if possible, update to the latest version of Pro.) ... View more

Hello Edvard. This should be doable.  I think the best way is to find all bus stops within the maximum travel time/distance limit of each origin and post-process the results. Configure the OD Cost Matrix so that it doesn't have a limit on the number of destinations to find, but use a limit for the travel time/distance.  (You could do three separate analyses for 300, 600, and 1000, or you could do it all in one using the maximum limit.)  This results in multiple rows in the Lines output for each origin point. Do some joins to transfer the frequency of service at each bus stop to the Lines output.  Exactly how you do this depends on whether you're using Pro and an NA layer or Python or ArcGIS Online.  Your question doesn't say. 🙂  I can help further if necessary and if you share some more information. Post-process the resulting table to identify the highest frequency stop within the desired distance limit of each origin.  This should be relatively easy using any type of database operation (SQL, Pandas, etc.).  You could probably do this with various Geoprocessing tools, too, like Sort and Delete Identical, maybe...? ... View more

Oh dear, I believe this is actually a bug in the tool.  I see that for some reason the travel mode parameter is a string-type parameter instead of a travel mode.  I honestly can't remember why I implemented it that way years ago, but it has the effect of preventing this very reasonable workflow from working. I think the fix should be straightforward but will require a bit of testing on my end.  Let me make sure I don't have anything else urgent to take care of today, and then I'll try to fix this for you. ... View more

Thankfully, this one has an easy solution! You just need to set  new_travel_mode.attributeParameters = attr_params after modifying the travel mode.  The TravelMode object doesn't allow you to edit the attribute parameters in place.  It won't pick up the changes. ... View more

The XML template is included in the Network Analyst tutorial data, which you can download from here: https://links.esri.com/NetworkAnalyst/TutorialData/Pro ... View more

This documentation page explains how points get snapped to the network: https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/locating-analysis-inputs.htm To summarize, the starting point of the analysis is the closest point on the closest network edge (or junction, depending on settings) that is not restricted.  The Service Area is calculated from that point.  The Service Area cutoff does not consider the distance from the network of the original point. ... View more

Bearing may impact which street an analysis input is located on (part of the locating calculation), but it does not influence the direction a vehicle arrives at and departs from a location (part of the solve calculation). For instance, if the vehicle's bearing points north, the Bearing and BearingTol fields will help it to locate on a north-south road. However, at solve time, the resulting route may travel south from the location to achieve the shortest path to the next stop. Use the CurbApproach field to control the direction of travel arriving at and departing from a location. In your example, 265 degrees is roughly east-west, so the point gets located on the east-west road (it also happens to be closest to that road and would have probably located there anyway).  If you used 0 degrees (north), it would likely have located on the north-south road, even though it's farther away. ... View more

They should get used automatically.  It just magically happens silently. Yes, it works in Pro as well as ArcMap.  Here's some documentation: https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/bearing-and-bearing-tolerance.htm  ... View more

I'm not completely sure, but it's probably because you're running the Model from the ModelBuilder canvas (using the Run button on the ModelBuilder ribbon), and the model is storing the intermediate outputs in the Table of Contents.  I think you have the "Add To Display" option turned on for one or more of the green bubble outputs from each tool.  That option is meant to be used for debugging, and it may overwrite things in the Table of Contents.  I don't really know what all the rules are. If you want to ensure consistent output, you should right-click the final output green bubble (from the Solve tool) and set it as a model parameter.  Open the model as a tool (not open for editing in the canvas) and run the model from the Geoprocessing pane.  Only the final output (the solved layer) will be added to the map. (Note from my response on your other post: I think you may have better luck contacting Esri Support.  You have a lot of questions, and they may be able to spend more time with you!) ... View more