RPA - Is the truth falling on deaf ears?
13 Mar 2018
Robotic Process Automation (RPA) is a clerical process automation technology using software robots or Artificial Intelligence (AI) workers instead of humans. It’s one of several emerging technologies brought about by the 4th Industrial Revolution.
The implications for organisations adopting RPA will be significant. But does it sound the death knell for industries that we thought would never see humans replaced by robots? Or will the opposite ring true for those who embrace this new technology as they grow organisations in new ways using the augmented data it offers?
RPA’s immediate benefit is that it will remove back office processing, and hand over manual data processing and other repetitive administration to a virtual team. This is an immediate cost-saver. High-volume repeat tasks in medical registration, invoicing and billing, payroll, accounting, finance, and procurement will no longer be managed by humans. It will enforce consistency and improve revenue, reducing costs by replacing the human workforce with something cheaper and more reliable. Customer satisfaction will improve through its speed and accuracy, and it will deploy software tools to do multiple digital tasks in an automated way.
RPA can do virtually all high volume business processing. In healthcare RPA can be used for:
- Receipt of e-referrals
- Patient registration
- Treatment eligibility and billing
- Updating patient records
- Generating outpatient referrals
How organisations can build on the benefits
As the range of benefits begins to flow, businesses will launch products to market faster, and the sharp-eyed early adopter will anticipate the problems this technology is likely to solve and get ahead of the game.
Roger Perry is Managing Director of the Bevington Group, and one of the region’s foremost productivity improvement and organisational design experts. He has advised organisations and government bodies on RPA and Artificial Intelligence (AI) and sees the healthcare industry adjusting to a world that few imagined possible just a decade ago.
“In healthcare, you can use robots to analyse data for clinical patterns and trends. You can ascertain when and under what conditions patients are most likely to be at risk, to get ill or worse. You can look at patterns for when you are likely at different times of the year to get caseloads coming into emergency. This means you can use data to help plan new resources and staffing levels as well as analyse it for clinical data. Using data to see patterns is something that is already happening in the medical field now, it’s used for diagnostic imaging for example, and it’s used for the logistical side of the business such as identifying when you’re going to be busy. It can be used for any administrative tasks or junior leadership activities that can be taught to an AI device. As long as the input and outputs you’re teaching them are correct they’ll form an algorithm. It can be everything from keying data, to paying bills, to preparing rosters, managing applications for leave and responding to simple payroll queries.”
As your brain absorbs the broader abstract playing field that RPA and AI is presenting, you can forgive yourself for foolishly thinking this was all about little mechanical beings marching up and down performing duties around the clock.
“There are also quite a few hospitals globally now using robots themselves, and not just robotic process automation” explains Perry. “Down the corridors of some of the global hospitals you’ll find little robots delivering clinically important items or messages.”
The same principles can be applied to education, where administrative activities are handled and improved by robots and through the use of AI. While a robot does manual data input tasks, AI carries out more sophisticated judgment tasks.
“One of the exciting developments for their use in education is for things like marking papers. In fact, it’s already underway.”
And if you think the idea of students attending a class led by a robot sounds silly, then Perry suggests that you think again.
“There are many experiments globally that seem to be bearing fruit. There are electronic tutors and lecturers¬–not humans but avatars. They have some wrinkles to iron out, but my goodness if you can get fifty per cent of the tutoring done by robots then you’ve massively reduced your costs. Better than that, if you can use your tutors and lecturers for the difficult tasks and for the simple tutoring activities use avatars and robots, you will provide the students with a much larger number of contact hours, and you can massively improve the educational outcome.”
RPA allows you to gather deeper insights from great volumes of data, and that helps businesses find new ways to grow. It provides the ability to manage multiple channels of information from various sources, managing its complexity and producing a rich bank of material for driving business development. It can do it quickly, too. With faster reporting times and data analytics that hold far greater value, not only will insights emerge but suggestions and scenarios will come into focus in ways that may hold profound meaning for growing revenue and for strategic planning.
Since the 1990s, many an industry has transitioned from traditional onshore labour to offshore to save costs. With 2020 just around the corner, we are now on the brink of a new revolution of digital labour that the industry has named “noshore.” But whatever you choose to call it, the characteristics of RPA are too important to ignore because they are self-learning, virtually autonomic, and self-healing (traits most humans find tricky to develop).
“The machine develops algorithms,” explains Perry, “a bit like algebra but more sophisticated, and it does so by watching for patterns in inputs and outputs. You give the inputs and tell the robot what the right answer is, and over time it will learn what the right answer is. They learn by developing algorithms.”
It’s a fine line to walk though, because the way algorithms work the better the data the more accurate the robot will be. At the same time, bad data going in can teach the robot the entirely wrong thing and it doesn’t have the sense to know that it’s wrong because it’s not human. But as with humans, no robot is perfect just yet and Perry points out there is room for redemption yet.
“Self-healing means that if an error is discovered, or processes are found to be sick, the robot can scan through, figure out what it’s done wrong and it can fix its own algorithm. You have to give these robots the right training. It’s not practically true yet that these are autonomic, self-learning and self-healing.”
For the healthcare industry, high volume transactional and data processing operations involve confidential and sensitive information, and risks in RPA adoption will need extra scrutiny.
The cost benefit means liberating humans to engage more deeply with providing services because costs savings can be invested in people. RPA will also strengthen diagnostic processes and help to improve hospital workflow and address many of the challenges around care coordination. You can also expect the evolution of Artificial Intelligence to become a growing trend creating greater demand for data mining. Why? It will support decision-making that produces better solutions in healthcare, as in other industries, because it offers an information advantage.
It's evolving now
Roger Perry sees a healthy adoption rate among organisations in Australia already.
“Lots of organisations are using RPA and AI in Australia. I was recently with people from the electricity industry, a distribution network that has lots of assets and equipment. The biggest cost they have is maintenance. How do they keep everything up and running so their grid doesn’t collapse? They usually do it with scheduled program maintenance, but now they can get an artificial intelligence device to shoot lots of photographs to figure out which poles are going to fail next. They put a drone in the air, fly over the infrastructure and take lots of photos, the AI analyses it all and you don’t need lots of people looking at the data. You can use software to figure out which poles are going to fail first and direct your assets there. It’s an enormous impact. It massively improves the resilience of the network, because you get to issues before there’s a fail, and you’re putting your resources where they are most likely to make a difference (in this case defending the network from impending failure rather than fixing something that doesn’t need to be fixed).”
Cyber security risks are very high and not just for the healthcare industry. There could be serious legal implications if problems arise using robots to manage internal business matters. How organisations develop compliance policies and take measures to protect themselves could mean the difference between business sustainability and being wiped out by litigation.
That said, the forecasts by some watching this space predict that the market for AI and cognitive computing systems in healthcare will achieve *revenues of more than $6 billion by 2021, a steep incline from nearly $640 million in 2014.
In 2018, you can expect more technology spend will go into digital, data, AI, and security, and this means squeezing IT spend on core legacy systems. Consumer-facing product demand and cybersecurity will therefore become critical issues for financial services, healthcare, insurance, and education.
But imagine what’s possible. Countries with a high population waiting in line at crowded public hospitals could get access to services more quickly, staff become liberated to recruit and train more health professionals, and medical specialists are empowered by deeper insights from data analytics when making their diagnoses. The impact on society, its insurance companies, fiscal bodies, pharmaceuticals, and a multifarious complex of suppliers in their chain is mind-boggling indeed.
The move to RPA and AI doesn’t necessarily mean high expense, and the opportunities for growing business in new ways could be lucrative for visionaries.
Mohit Sharma is an expert in helping organisations transition to automation and AI operations. His company Mindfields has helped global banks navigate the process.
“When large amounts of core processes and activities become automated by software, the software itself becomes the differentiator between the performance of rivals in any given industry.”
This signals the importance of making good decisions from the outset to support business sustainability and growth, post-transition.
So how can your organisation be sure it has the right BOTS?
“I don’t think it’s as much about the RPA solution than finding the right RPA partner,” says Sharma. “People need to realise it’s not about the software, there are lots of different types of RPA and AI software out there and actually the software is often quite inexpensive.”
Australian companies may be disadvantaged however; network capabilities in many other countries provide faster broadband speeds than we have at home. It raises the conundrum of whether or not this places pressure on Australian companies in their ability to compete in a global business context, but both Sharma and Perry see this as a minor setback.
“It’s not a big problem. Australia has a small population and we don’t require that much speed,” explains Sharma. “It is a factor, but it depends which industry and which process you’re talking about. For example, if we are using a TPG 400 MBPS connection we can serve our client faster than somebody using an old ADSL2. In Australia, speed is not the only differentiating factor, it’s one of the factors. Privacy, security, and knowing what the consumer wants are important factors. How to apply technology in ways that align with these factors is what matters."
"If we look at this in another way, there is also the coding. How you make a program in technology is also a key differentiator. Taken from another perspective, India is providing the biggest number of developers to the world and yet they are the least innovative developers.
Having the skill set of coding will not necessarily give you the best software. It is how you apply it that’s critical. Speed is only one small factor. How you apply the knowledge to the whole process is much more important.”
It’s an outlook that Perry shares. Both experts think innovators in industries will recognise the opportunities that augmented information brings to the table, and proactively step into RPA and AI if they haven’t already done so. If you can look beyond work displacement to meaningful interaction, and foresee humans undertaking tasks of greater value in your organization, then you open the door to opportunities for business growth. In healthcare, success may appear quickly in ways that are obvious, providing more human contact rather than less in an increasingly social work environment. The gains in improved business performance are surely a driver for creating revenue and this principle applies to all industries, because happier and more collaborative humans can really turn up the volume on productivity.
Roger Perry is the Managing Director of the Bevington Group, and one of the region’s foremost productivity improvement and organisational design experts. He has been an Assignment Director and Steering Committee member on over 40 transformation programs. He advises the region’s leading businesses and government entities on operating model design and digital strategy.
Mohit Sharma is Mindfields founder and Managing Director. With over 20 years of experience in Strategy, Corporate Finance and Risk Management solutions, Mohit understands the drivers for change that most businesses face. Recognised as an industry thought-leader in Robotic Process Automation, Artificial Intelligence and Outsourcing, Mohit is regularly featured in leading financial publications and media for his insights into the ever-changing RPA landscape.
Smith, P. (2017), Robotic process automation on demand as consultants get disrupted, Financial Review
(2016), AI systems market for healthcare to expand from $600 M to $6 B, Healthcare innovation
Robotic Process Automation – friend or foe for your risk profile? PWC