Robotic Process Automation

Introduction

RPA is officially dead. Please refer to Intelligent Process Automation or hyperautomation

What is RPA?

It's a class of software that allows you to transact in any IT application or website typically in the same way a human would, to perform complex rule-based work.

Automation means not having to do anything twice. Automation with the Internet means not having to do anything that anyone else has already done. All that remains are creativity and judgement. — Naval

RPA is a simple technology that completes repetitive, rule-based actions from structured digital data inputs. RPA automates processes and tasks by mimicking action through scripting and following rules. In contrast, cognitive automation leverages learning, reasoning, and self-correction.
– Theo PiletskiOpens a new window, Vice President, Business & Strategy Consulting, EPAM Systems, Inc

The right model to think about productivity in software engineering is to consider that each engineer is backed by an infinite army of robots.

View more quotes…

In augmented work environments, you generally see a positive impact on performance. Increased efficiency in work leads to increased output, which in turn pushes up performance indicators. People are also likely to report greater job satisfaction because their mundane work has been taken on by RPA systems and their freed time is allocated to other types of work that are more enjoyable. — Rohit Talwar, AI Expert, Fast Future Research

I would love to be paid purely for my judgment, not for any work. I want a robot, capital, or computer to do the work, but I want to be paid for my judgment. — Navalism

The human touch is indispensable to creativity and managing the unpredictable. — Valérie Claude-Gaudillat, Audencia business school

Every person’s job is likely to be changed by at least 25 per cent over the next decade. — Leslie Willcocks, LSE

Far from taking over, Automation will most likely help us cope with the dramatic and under-researched intensification in the pace of work. — Leslie Willcocks, LSE

For me the most interesting question is not: ‘how do we preserve what's human about our work?’ It’s how do we ask the right questions about what trade-offs we are willing to make. The idea that automation is better than humans across the board is incorrect. What we need to do, is find out where it can really make a different for the better. — Dr. Genevieve Bell, Australian National University

Possibilities do not merely add up; they multiply. — Paul Romer, co-recipient, 2018 Nobel Memorial Prize in Economic Sciences

We are transitioning from our initial focus on the financial benefits of the technology and are now building the business case to get the technology into every aspect of the organization. We can say, ‘look, here is an opportunity to unleash creativity in all areas of the organisation and maybe people who were not previously in a position to innovate will have the capability to do that. — Amory Booher, Senior Vice President, Risk Technology and Productivity, BBVA Compass

Think of the human body breathing. It’s a complex and critical mechanism, but automated, so our brains are freed to power everything else we do. I think for many organizations, all they can do is breathe. It’s so important, it’s all the employees can focus on. But when that breathing is automated within the organisation, then employees can focus on so many strategic issues and opportunities. — Mihir Shukla, CEO, Automation Anywhere

A lot of the benefits that come from augmentation don’t just result from the augmentation itself but from the general reappraisal of how the organisation is operating. — Ben Hammersley, Futurist

Technology is a useful servant but a dangerous master. — Christian Louis Lange

Without change there is no innovation, creativity, or incentive for improvement. Those who initiate change will have a better opportunity to manage the change that is inevitable. — William Pollard

Robots and computers will own any clearly bounded job, skill or game. Humans should focus on creative work. — Naval

Organisational models will need to adapt as more administrative tasks are taken on by smart software and automated processes. We may even see businesses rethink their decision to offshore shared services given that if they can rely on virtual assistants to handle more day-to-day processes there is less opportunity for cost arbitrage. — Chris Hogan, Head of Global Client Relations at CIMA

Useful technologies to learn for RPA developers:

RPA Development: Java/C, JavaScript, Ruby on Rails, XML, Dotnet.
API Development: C++, VB6, C#, VB .net, and Delphi
Advanced business user: programming logic like If/Else, List/Array/Variable creation and use, basic database queries.

Rule-based actions

RPA is a versatile form of software automation that allows businesses to relieve teams from many of the repetitive, rule-based actions, and processes that might otherwise exist as pinpoints within the operation.

Rules-based principles explained by Isaac Asimov's in 1965 (the three laws of robotics)

If they can train you to do it, then eventually they will train a computer to do it.

Isaac Asimov's "Three Laws of Robotics:

A robot may not injure a human being or, through inaction, allow a human being to come to harm.

A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.

A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

Download as image

Infographics

↑ Which Professions Are Most Vulnerable to Automation?

↑ What processes should we automate for RPA?

↑ Automation Architecture example

↑ Which Productivity gains from RPA alone

Start Off Simple Automation is a bit like digging a tunnel: You don't know what you're going to run into. It's best to begin with something useful but simple. You'll avoid wasted time and resources but learn a lot about your organisation and infrastructure. Carry those lessons forward, taking an incremental approach, and you'll maximize your productivity gains while preparing yourself for bigger projects. — Matthew Russell, Digital Reasoning.

Different types of automation

Key differences	RPA	IPA	Hyperautomation
Tools or technologies required	Performed by RPA automation tool	Performed by multiple machine learning modules and automation tools. Custom development required	Performed by multiple machine learning, packaged software (Process Mining, No Low code apps, analytics) & automation tools
Maturity level	Very matured, sophisticated, fast and easy to implement	Scaling. Sophisticated AI-based process automation with cognitive ability	Transforming. Sophisticated AI-based process automation with cognitive ability and can loop humans in the process
Outcome	Efficient operations	Efficient complex operations	Smart and efficient operations
Scope / Coverage	Rules based, routine repeat mundane tasks	Higher-function tasks that require some level of reasoning, judgment, decision, and analysis	All-ecompassing: "everything that can be automated will be automated"
Governance approach	Process-first	Process-first	People-first and Process-first
Implemented by	IT, citizen developers	IT	IT, democratization of Automation Development

Note: Robotic Desktop Automation (RDA) is configured software that works in real time with a human operator by presenting a predefined set of information coming from different systems to assist the operator with the completion of operator-generated workflows or tasks.

Vendors

Leading RPA Vendors:

Other vendors

I also encourage you to check out other resources, such as the works of professors Mary Lacity and Leslie Willcocks. Publicly available research on RPA by leading analyst firms such as HFS Research, The Everest Group, The Gartner Group, Nelson Hall, and The Forrester Research. Many RPA vendors have online forums and some, as I mentioned, have outstanding academies. A.J. Hanna, has a YouTube channel called Ask A.J. in which he shares is own experiences from having been an early adapter of RPA.

Differences between RPA and other tools

Let's wrap this chapter up quickly by addressing some of the most commonly asked questions I've seen in the market because I'm pretty certain you're either thinking them or you'll soon be asked them. The first is, isn't RPA the same as screen scraping or just macros on steroids? The second is, how is RPA different from business process management tools? And finally, one I'm certain you'll encounter, how is RPA different from artificial intelligence? The macros on steroids description has been around for a while and even though it sounds negative, I think it's actually somewhat helpful.

BPA	RPA
An approach to optimizing business processes for maximum efficiency	A software that automates and optimizes a certain business process
Automating processes across an organization	Specific, task-oriented automations
Requires in-depth analysis of business problems	Can operate within existing business processes
Requires building a solution from ground up	Elmininates tasks by performing them just as a human would

First, because people understand what a macro is. When presented with a task they need to repeat over and over again in an application like Microsoft Excel, they build a macro to emulate the actions they've been taking with their keyboard or mouse. So to that extent, RPA bears a resemblance to macros, but they can do a heck of a lot more, hence the on steroids. RPA is able to control a massive catalog of applications and integrate between them. And most RPA tools have out-of-the-box capabilities to handle complex logic, calculations, and sophisticated workflows.

So yes, to some degree, RPA is like macros on steroids, but if capability, control, and security are important, I'd choose RPA every time. As for screen scraping, hopefully it's clear that the sophistication I had just described also highlights that RPA can do quite a bit more than simple scraping of data from screens. Next, we have business process management tools. BPM isn't a specific software tool, but more of an approach to streamlining business processes to create efficiency and value.

It's a deep look at how processes are operating, highlighting areas for improvement and building solutions usually from the ground up. RPA, on the other hand, is designed to operate processes as a human would so it exists in more of a surface level. As we've already covered, it's often faster to implement, works with almost any software, and is easily adapted to the changing application environment. RPA and BPM are not necessarily in conflict with one another. Instead, they share the same goal, just with different implementation strategies.

Finally, artificial intelligence. This is a huge topic. For the sake of this article, I'll try to be concise. AI is an umbrella term. It's used to describe a spectrum of capabilities that includes machine learning, natural language processing, complex pattern recognition, hypothesis generation, and the iteration of algorithms as they learn from more data. Think of the Netflix Show Recommendation Algorithm for example which learns from your browsing habits over time.

You may also hear the term cognitive in your research. It's a subset of AI and the term has been applied to tools we've long known as IVR, Voice Recognition, and OCR, Character Recognition. It's these cognitive tools you'll most often encounter especially as supplements to the RPA capabilities you'll get out of the box from most leading vendors. RPA is not typically labeled as cognitive or AI because the underlying technology is different. With RPA, a developer creates a rule-based process that the robot fully conforms to and follows.

You saw examples of this in the previous lesson. The rules won't be adjusted unless a developer reprograms them. This means that everything RPA does from beginning to end is completely dependent on how it was configured. So RPA by itself cannot be considered intelligent. And while it prompts some to call RPA robots dumb, I can't disagree more. They're just well-behaved and it's their consistency that makes them so incredibly valuable. Plus, the well-behaved RPA is only just the beginning.

As you get more advanced, you'll find you can combine RPA with specialised cognitive tools to achieve intelligent digital labor. For example, RPA lays a foundation for other tools down the road such as machine learning by creating copious amounts of transactional data to feed the machine learning algorithms. Having the proper outlook can allow you to proactively accommodate these technologies when they're enterprise ready, fit to be part of your future of work toolkit. So remember, RPA is not macros on steroids, it's like it but much more capable.

It's a complement, but not a replacement to BPM. And RPA is not the same as AI, but it will serve as an invaluable precursor to applying cognitive tools to your organisation. I'm often telling enterprises that to build your cognitive castle, you must first lay a digital foundation. RPA is that digital foundation on which you can employ more advanced technologies when the time is right and if the solution requires.