RPA Orchestration, The Killer Use Case
Over a year ago I wrote a blog entitled, “RPA 2018: Many Bots Make Light Work”. I tackled the Orchestration side of RPA and talked about how UiPath Orchestrator allowed RPA to do something it hadn’t been able to effectively previously, SCALE!
RPA started as a back-office scripting solution. One would have to finely tune every machine a script ran on. Moreover, a developer had to copy the code from one machine to the next and so on. In addition, there wasn’t an easy way to manage the automations centrally. Sure there were some “cobbled” together solutions using shared drives. However, there was no way to employ “command and control” capabilities from a centralized workstation.
As I mentioned in other blogs I created a utility similar to the Orchestrator but that was back in 2015-2017. Technology has certainly moved forward since that time.
This year UiPath has made a push to emphasize their Enterprise capabilities. Two products were created to put them in that direction, the Connect Enterprise Hub application which gives an organization a centralized area to vet automation candidates, manage projects, create documentation and integrate with the Orchestrator. The other product that adds to UiPath’s is their Process Explorer tool that can be spun up to analyze desktop processes. Their recent purchase of Process Gold should only enhance these capabilities.
To be clear these additions bring incredible value to UiPath suite but… the Orchestrator is the backbone to everything in the UiPath world.
The good news is that they haven’t stopped adding more features as evidence by their latest 19.10 release. One of highlights include “Queue Triggers” which means jobs can now start when queue items show up in the queue. The Orchestrator then pushes the information to any machine that is ready to take the request. This is a huge boost in regards to scaling. The big plus is that you don’t need have a job constantly scheduled looking for queue items. You also don’t need to have a bot sitting out there 24/7 waiting for requests. This command and control capability is very powerful.
That being said, I’d like to shift focus to the emerging set of AI and ML tools.
As Craig Le Clair stated at Forward III this year, the one thing that that AI/ML Vendors don’t have is… “Orchestratration”. This is very meaningful and not something to be taken lightly. In addition to being a top shelf RPA/AI engine the UiPath Studio is above all else an integration engine. The Studio’s architecture that allows a developer to either utilize, download or create custom activities. This unleashes unlimited capabilities of what can be accomplished. The abstraction and the organization built into the UiPath framework increases its reliability.
Taking these factors into account, one can create a package that includes RPA, ML, AI or a code block, then compile and deploy it to the orchestrator. The orchestrator can push it out to machines to run. It provides a birds eye view of the running bot. The bot itself is agnostic to what it’s running since it’s essentially a compiled program with references to a range of assemblies.
I realize I may have gotten a bit technical there. However, I feel it’s important to point out how this fundamentally works.
To put it in simple logic, you can create anything within he UiPath Studio and utilize “best of breed” activities. Once the work is done there, the Orchestrator can then run this package on any machine has that has been provisioned. In addition to logging and other job specific activities, the Robot can be scheduled, started/stopped etc.. While this isn’t a complete assessment of it’s abilities, it only touches on the capabilities the UiPath Orchestrator can provide an enterprise regardless of what type of job it executes.
Peter S. Camp is the CTO and Founder of CampTek Software. He has been developing RPA Applications for over 15 years. For further questions, discussion or inquiry about CampTek Software Services, contact email@example.com.