Sketch… | Personal Opinion

Automation is the sibling of digitization. With the recent great resignation and labor shortages, it’s opportune for us to reflect on the topic of automation.

Procedural Automation

Any informational tasks carried out in a sequence of steps is amenable to automation.

UI vs API

Any information system must provide its affordances in forms of User Interface (UI) or Application Programming Interface (API) which are for human and programs respectively. UI can be further categorized as Command Line Interface (CLI) or Graphical User Interface (GUI).

Based on whether the host app provides an API, automation can be done with simulated point-n-click robots or programmatic scripts.

Robotic Process Automation (RPA)

Many legacy systems provide browser based GUI but no API. To automate tasks carried out in these systems, robotic process automation can be used. Lifted right out of one of the RPA vendor’s home page: “Robotic process automation (RPA) is a software technology that makes it easy to build, deploy, and manage software robots that emulate humans actions interacting with digital systems and software. Just like people, software robots can do things like understand what’s on a screen, complete the right keystrokes, navigate systems, identify and extract data, and perform a wide range of defined actions. But software robots can do it faster and more consistently than people, without the need to get up and stretch or take a coffee break.” In a word, RPA automates manual steps carried out within an UI.

Important thing to note about RPA is, it’s only suitable for digital systems without an API. Applying or selling RPA on top of system with an programmable API is, how should I put it, pure fraud! Mark my words: RPA vendors will eventually go extinct as more and more digital systems provides API. When that time comes, you may find RPA vendors forced to retreat to unexpected area such as automatic GUI testing.

Scripted Automation

For digital systems with an API, automation should be carried out by programed scripts.

Decision Automation

In a traditional analytics workflow, analysts summarize data into reports or dashboards which aide decision makers in their decision-making process. The process is manual and human-in-the-loop.

With the advent of AI/ML models, many decisions can be automatically made, without people-in-the-loop.

Semantic Automation

Cognitive functions which are traditionally out of reach for automation are now automatable with technologies such as CNN, speech to text and NLP/NLU…

Chat Bot: an interface to bring them all together

To be finished…

References:

  1. The future of automation, a semantic perspective.
  2. What’s RPA?
Automation