In this article, Lewis Silkin LLP looks at what RPA (robotic process automation) is, the impact on the future world of work, and how it might affect employment law.

How RPA works

The influence of automation in the workplace, both now and in the future, is undeniable. From manufacturing and shipping to financial services and healthcare, automation is at the forefront of efforts to speed up and refine work processes. Research from The World Economic Forum indicates that, by 2018, machines were already responsible for 29% of total task hours in the workplace – a figure expected to reach 42% by 2022. This article focusses on Robotic Process Automation and the continued impact it stands to make on how we work.

How Robotic Process Automation works

Once an organisation identifies which of its tasks or processes are suitable for automation, it can set about implementation. The technology consists of software robots that are able to replicate a human worker. These robots, once programmed with instructions, can log into applications, enter data, calculate and complete tasks and then log out. RPA technologies are broadly seen to be made up of three categories:

  • Probots – following simple, repeatable rules to process data

  • Knowbots – searching the internet to gather and store user-specified information

  • Chatbots – acting as virtual agents or virtual assistants who are able to respond to customer queries in real-time

RPA software is not part of an organisation's IT infrastructure. Instead, it sits on top of it, enabling a company to implement the technology quickly and efficiently - all without changing existing infrastructure and systems.

Is Robotic Process Automation AI?

In practice, RPA mimics human actions whereas AI mimics human thought. RPA handles easier, mainly physical, tasks that require no knowledge or understanding - they can be done by simply codifying rules and instructing the software to act accordingly. AI involves dealing with natural language, reasoning, and judgement where the machine learns on the job in the same way as a human being would do. Importantly, the two technologies are increasingly being used in tandem to improve outcomes for businesses.

Robotic Process Automation in HR

One of the areas of business operations most likely to benefit from RPA is human resources. Automation can facilitate ‘onboarding’ – a key function involving all the steps needed to get a new employee deployed and up to speed. For example, an automated process could be triggered by a manager updating the status of an applicant’s file to ‘hired’. This would initiate the delivery of application forms and CVs to the right departments for archiving, the dispatch of a job offer letter to the applicant, as well as the creation of user accounts on the organisation’s IT systems and the generation of an ID card. The same concepts are, of course, applicable to the reverse function of ‘offboarding’ departing employees.

As far as such mundane tasks can be automated, HR professionals should have greater freedom to engage with the more creative and complex sides of their roles. After all, some aspects of the workload, such as handling interpersonal or disciplinary issues, still require a human approach which is hard to replicate through automation.

Benefits of Robotic Process Automation

Opportunities to undertake more varied and interesting work could be welcomed by employees who typically dislike the jobs which RPA can now cover. Additionally, the Trades Union Congress has argued that increased use of technology, including RPA, could lead to a reduction in the number of people needed for work and eventually result in a four-day working week.

As far as employers are concerned, the commercial incentives for introducing RPA are compelling. The cost of acquiring a robot licence is typically lower than human salaries and RPA is reported to provide between 30% and 200% return on investment after the first year of implementation. Over and above direct economic benefits, robotic software, unlike humans, can operate 24/7 without becoming sick or needing a break. Organisations can increase their workloads whilst reducing the time taken to get through them.

Robotic Process Automation examples

RPA has huge potential in almost all industries but stakeholders in insurance, financial services and healthcare are especially well placed.

RPA and insurance

Automation can assist in generating renewal premiums, processing claims and policy administration. Zurich Insurance took advantage of RPA technology to improve policy issuance processes. In collaboration with digital transformation company Capgemini, Zurich developed an approach to RPA whereby its standard international insurance programs could be issued by a software robot. This freed up the organisation’s human workers to focus on more complex policies as well as potentially improving customer service and reducing operational risk.

RPA and financial services

Key functions like account openings and closings, management of audit requests and mortgage processing can be enhanced by automation. The Singaporean bank OCBC reportedly succeeded in using RPA to reduce the time taken to re-price mortgages from 45 minutes to only 1 minute. Here, RPA was used to check the customers’ eligibility for a review of their mortgage, recommend new price options, and ultimately draft the recommendation email.

RPA and healthcare

RPA can facilitate decision making on prescriptions, the management of patient records and patient registration. University Hospitals Birmingham trialled the technology by introducing self-service registration kiosks. The result was a doubling of patient flow along with a 50% improvement in staff efficiency.

RPA vs Intelligent Process Automation (IPA)

Whilst RPA robots improve the carrying out of repetitive tasks, they simply do what they are told to do and therefore cannot adapt well to new parameters. IPA builds on RPA by creating learning systems that simulate human behaviour.

For example, companies often use service tickets consisting of unstructured data regarding error notifications. These are then compared to a defined matching list. Once an error description matches one of the stored keywords, it can be automatically allocated, processed and then completed, without human interaction. However, when there is no matching for an error description, traditional RPA reaches its limitation. The process cannot be automatically completed and gets rejected for a manual take-over. Human experts then classify the rejected process and assign it to a keyword from the matching list, allowing the case to be processed and completed. IPA precisely analyses these interactive rejections and can then carry out future classifications itself. Step by step, this leads to a decrease of rejected processes and simultaneously increases the level of automation.

What is the future of RPA?

In the long term, RPA looks set to feature more heavily in the workplace. Applying the technology will be necessary as companies, and indeed countries, try to remain competitive. The UK is far behind many others in this respect - a recent BEIS report pointing out that, in 2015, the UK had just 10 robots for every million hours worked as compared to 167 in Japan - and should take a bigger part in the so-called ‘Fourth Industrial Revolution’ sooner rather than later.

However, implementing RPA requires preparation and a clear strategy. Unless existing workforces are educated on how to work with the software there is little use putting it in place. Organisations, like Zurich, which have committed to RPA have gradually built centres of excellence covering skill development and capabilities. These centres include people who assess the viability of a new RPA project as well as those responsible for configuring, installing and developing the robot. Others then plan the work of the software and how to fit it in with human work.

Adopting such a structured approach requires those in management positions to understand the benefits of RPA and convince employees, some of whom will fear for their jobs, to see them too. It may be that the rate of automation in the workplace ends up being slightly slower than expected.

Robotic Process Automation and employment law

In the short term, the impact of the increasing adoption of RPA on the workforce may be low. However, the impact will more likely be felt in the longer term, as the technology matures and has an increasingly transformative impact on jobs.  While there is no clear consensus on the impact of RPA on jobs, it seems likely that it will have a significant impact on jobs at the task level rather than eliminating entire jobs.

Implications for HR

With the potential for RPA to deliver opportunities for employees and productivity gains for employers, HR will have a significant role to play in reshaping the workforce to harness these potential gains. As work evolves to incorporate RPA, HR will need to consider the impact of these new technologies on the workforce and develop strategies to support employees through retraining and deploying individuals to focus on higher-value or more complex tasks. Workforce anxiety around the implementation of technology and fears of redundancy should not be underestimated, particularly for workers in certain job types who are more vulnerable to being displaced.  Adopting a culture of collaboration, adaptability and lifelong learning will be a significant enabler in the successful adoption of RPA. This will need to be supported by effective leadership and communication about the impact of the new technology on the workforce.  Indeed, some organisations may have to consult with employees about the use of labour saving technology under European Works Council regulations.

Re-training and redundancy concerns

With this in mind, the UK government’s national retraining scheme seeks to enable people to work with, rather than be replaced by, machines - a new pilot scheme will usher in an extra 2500 AI and data science conversion degrees. In the longer term, we may see an increasing focus on existing redundancy laws – for example, to ensure that employees receive retraining where roles become redundant through technological disruption. A reduction in the qualifying period for redundancy pay (currently two years’ service), or an increase to statutory redundancy pay entitlements, could drive employers to invest more heavily in retraining in these circumstances than is currently the case. With ongoing concerns about rising insecurity at work and the impact of future technological developments on employment, the current debate on the benefits (or otherwise) of introducing a universal basic income will no doubt continue.

Employers’ duty of care

Automation can help to enhance wellbeing within the workplace itself. Recent campaigns, such as #thisplaceminds, have highlighted the duty of care upon employers towards their employees as well as the prevalence of stress, anxiety and mental health issues in the workplace. Where automation could release employees from unsustainable workloads, or even provide more certainty and transparency around grievance processes and general safeguards in the workplace, employers should consider implementing RPA. Conversely, employers should be mindful of the speed of their automation adoption and the impact on the workforce - employee attitudes, and those of  wider society are changing, with increasing expectations of ethical adoption of technology in the workplace – this evolution of the employment ‘deal’ will feature in our Future of Work Hub events (for more information click here).

Comment on RPA

Change to well established modes of work can be disruptive and, particularly in the UK, organisations must consider whether they have the infrastructure to adopt RPA. However, fear of the unfamiliar should not be a barrier to taking advantage of all that the technology has to offer. With greater job satisfaction and cost savings on the horizon for employees and employers respectively, RPA is set to play an increasingly significant role in shaping the future of work.

By Lewis Silkin LLP


Further resources about RPA (robotic process automation)

Comment