Flagging inspected assets with ArcGIS Collector

[TahirTamba1]

Hi,

I have several assets to inspect in a small area. How do I know if an asset has already been inspected so that I cannot inspect the same asset a second and third time. Is it possible to configure the Collector form app so that when the form is completed and submitted by the field worker, the point changes color according to the predefined color?

The only working alternative is the join between the feature layer and the inspection table by creating a style based on a single symbol using location.

However, this approach limits my operations since the inspections are recurrent and are done on a weekly basis. When the current round is completed, I have to truncate the inspection table for the next round.
This approach is time consuming and I have to backup and archive the inspection table. But also, it is inefficient to produce a dashboard from inspections made with ArcGIS Collector.

Thank you!

[ttamba]

Hi Jonathan,
Thanks for your feed back to my post! I thought I was the only person who had this difficulty in inspecting assets, because so far you are the second person who seems to have the same problem as me.

I had submitted this question to ESRI Canada technical support and a specialized analysis in Arcade confirmed to me that Arcade is the language that could solve the problem. He accompanied me in the implementation of the code that I posted in my first post. However, the tests I performed allow generating a unique symbology regardless of the date and time ...

The inspection of municipal assets is a recurring operation in the public works departments of cities around the world. Such operations are carried out every day in a repetitive manner on thousands of assets. For example, in the case of fire hydrants, field workers inspect them several times. Without a visual method of representing the inspected asset, the risk of errors in inspecting the same asset on the same day or on subsequent days in the same round is very high.

Indeed, I agree with you that ESRI must integrate in its roadmap a future implementation to flag an inspection carried out through a color or symbol change of the point in its Field Maps or Survey123 mobile applications.

Fortunately for us, the only sticky option is the join between the hosted feature layer and the inspection table so that we can create a unique location-based symbology every time the form is submitted.

But Unfortunately, this option has limitations, because after inspecting all the assets, it is necessary to empty the layer and recreate a new join for the next rounds. An eternal restart!

I also put the idea of ​​implementing the change of color or symbol to the vote in the section of ideas suggested by users for Fied Maps, three weeks ago. At the moment, there doesn't seem to be a buy-in to this idea, but I'm glad to see that I'm not alone on this adventure.

Kind regards,

[JonathanMcDougall]

@ttamba 

Hi, you're most definitely not alone.

A solution seems to tantalisingly close! I really want my organisation to rationalise its asset management systems; highway, land & assets, tree, and other pieces of infrastructure inspections. It's the big volume items and that visualisation of a completed inspection that's almost pivotal to a complete move towards ESRI.
Over the last year or so I've thought of "What ifs". For example; on selection for a new inspection the coordinates of an asset are passed to the inspection table thus allowing the table point to be drawn and symbolised. Likewise on the reverse, an "inspected" flag is pushed from the table to the asset which would then just mean the reset is a simple calculate field on the asset to reissue the inspections.

If I had the coding skills, I'm sure I could get closer but this has not been my background - was a field grunt for most of my days.

Perhaps, with the new map viewer and Field Maps Esri are looking towards something that would give us lesser mortals a solution.

[ttamba]

Hi Jonathan,
Knowing that I am not alone comforts me, but also knowing that the issue raised is a concern for ESRI as there are looking for a solution.
I also appreciate the way you approach this issue and prove that you have a better understanding of the issue even if you are not a coder. Your approach is logical and clear... I would like to be a code guy to develop this solution, unfortunately that's not the case for me also...
Like you, I have not yet made the jump to the new version of the map viewer and the official version of Field Maps. But I hope that Field Maps has more advanced features than its predecessor Collector and some improvements that will make our life easier.

Regards,

[Kevin_McIntyre · Accepted solution]

What you specifically want from ESRI is to be able to use the FeatureSetByName() function in Arcade to create a symbology expression. AGOL and FieldMaps do not support this function for whatever reason when using it for symbology. So while you could create a popup expression in Arcade to pull from a related table and display the most recent inspection in the popup, you won't be able to symbolize your features based on that expression in the popup. 

 As far as not knowing how to code, ESRI did a pretty good write up on using web hooks in AGOL with no coding involved and if you spend some time reading up on editing features through the REST service, you could then update that Inspection flag field in your main feature class to show the last time the feature was inspected or is due for an inspection.

Unfortunately, ArcGIS Online is in my opinion a service that is to be developed upon for organizations to tailor their own needs to. While there are a lot of features packed in AGOL that make it user-friendly and useful in some organizations, it is fairly bare bones as far as asset management software goes and requires a fair amount of technical expertise to go beyond simple feature edits/creates or displaying of data. In my case, I had to develop an entire Python web application that could receive web hooks from AGOL and manipulate the data to do what I needed it to, including flagging assets for inspections and automated work order creation in WorkForce. A simpler version of this would be to create a scheduled Notebook to run every so often and update that InspectionFlag field in your feature class based on the most recent inspection.

[JonathanMcDougall]

@Kevin_McIntyre 

Thank you for taking the time to respond, and thank you for the link too.

For many years my background was using, then managing, Pitney Bowe's Confirm asset management system, so you can perhaps appreciate why I'm looking for that little bit more from AGOL, especially as I can see the real opportunities within the platform.

We have Notebooks in the cross-hairs to examine how we can do some programmed dataset housekeeping bringing in some background automation. I just need to free up some "after work" time and start my journey. We're also just moving to an Enterprise architecture, so hopefully will also be able to bring in some of the FME processes I run elsewhere.

Keep an eye out for more questions, it'll be obvious when I start looking a little deeper into things 🙂

[Kevin_McIntyre]

@JonathanMcDougall 

Notebooks will definitely help you out with automating some of the more mundane data tasks at first, and would be a good introduction to using Python with your GIS. 

However, notebooks will only get you so far, and if a full-fledged asset management experience is what you want, you'll want to switch on over to using Webhooks to automate your services but I'm also aware that ArcGIS Enterprise may have extra features that help with automation without having to learn to code.

If you are interested in learning to code, I've used a website called codecademy that was pretty good at teaching basic Python syntax, but probably any code education website will be more than adequate... especially considering Python is probably one of the most popular languages to learn right now. 

Once you learn how to code, you can implement something similar to what I have setup. What I have going is a Flask application that receives the Webhooks, and then I used a factory design pattern to serialize the data and push it to the proper functions to manipulate the data in the feature service. Doing it this way allows the code to be very modular, adding in new features is wicked easy, and it can support multiple feature services.

It also wouldn't be a bad idea to ask your employer if they'll pay for your training or to let you learn on the clock. It's well worth it considering you'd be making your organization/business more efficient. 

[JonathanMcDougall]

@Kevin_McIntyre 

We're not a big local government authority here in Scotland, so our asset management data needs and aspirations are not huge. If we can get to a point in the near future where more of our services are taking a data driven approach, I'll be happy. Even happier that I'll be moving them towards improved data standards, interoperability, reuse, transparency etc..

Again, thanks for the advice - hearing experiences are infinitely useful.  I'm putting together a Data, Research, and Innovation playbook within ExpBuilder and have just added a Code section to "bookmark"  some of your info.

Point taken re training, however, I usually do a little bit of homework first to have an evidence based approach to proposals. If people can see some tangible benefits they are easier to "manipulate".