In the face of an increasingly aging society, how shall we best serve the public in a fast way in dozens of business scenarios like national pensions, child allowances, disability benefits, nursing insurance, household registration management, and housing registration?

Instead of simply adding manpower or equipments, or increasing service time, the U.S. and Japanese governments are adopting an increasingly popular technology—RPA, Robotic Process Automation.

In the United States and Japan, the most popular scenarios in RPA adoption may not be enterprise services as we imagine, such as Smart Finance, Smart HR, etc., but the public services and civil affairs. Through RPA, they have increased operational efficiency, and reduced labor costs, which has seen an unprecedented improvement the input-output ratio.

Globally, the public sector employs about 33% of the global workforce, and consumes about 30% of global GDP. So RPA has been increasingly favored due to its satisfactory results.

As the next step, the United States and Japan is planning to combine AI, deep learning, BPM, and other technologies to expand the RPA scenarios with the aim to improve the experience in public services and civil affairs.

The Haybie Institute believes that as the digital transformation has become the most pressing need of enterprises, RPA is considered to be one of the best tools for them to realize it. RPA+AI upskills the employees in enterprises, and boost their productivity in business activities..

01

What does public services do with RPA?

So what is RPA? To put it simple, it refers to the robot configured to automatically perform tasks, such as automatic data entry, uploading, downloading, summarizing, analyzing, etc. Behind these tasks is the technology called RPA.

RPA automates rule-based, repetitive, and tedious digital business processes by simulating human operations, to reduce costs and increase efficiency. Therefore, RPA is said to be the simplest AI application.

In Japan, RPA has been widely adopted in government agencies in the 45 provinces to address challenges posed by the aging society.

The Ministry of Internal Affairs and Communications of the Japanese government had ever released the report on the AI·RPA application for two consecutive year. Through the thorough investigation on 1,788 government agencies, the Ministry had its 2022 research report available. Its key findings include:

RPA is widely used in government agencies in Japan's 45 prefectures. Among them, government affairs ranks top in RPA adoption, which can be seen in 265 agencies; personal income tax the second, in 225 agencies; health management the third, in  175 agencies.

Its adoption involves many scenarios, which has yielded good results. They mainly include financial accounting, education, fire prevention, document management, child allowances, nursing insurance, fixed asset tax, as well as state pensions, household registration management, housing registration, disability benefits, etc.

Finance, accounting, and financial services are top scenarios in AI+RPA adoption. The same is true to the RPA adoption, which involves 306 instances; the child care and nursing services follow, involving 177 instances; administrative services involving 175, and other services involving 335.

AI is mainly used in text recognition, speech recognition, image recognition, numerical judgment, and other services. For example, recognizing information on images through OCR. Through investigation, we have concluded that 23 cases are for voice recognition and 275 for text recognition.

Due to their complementary nature in technical adoption, RPA can simulate human behaviors like reading, speaking, listening and so on with the help of AI. So most business processes using RPA can be enhanced with various AI capabilities..

The U.S. federal government shared the role of RPA in the digital transformation of the government in its two RPA Application Reports released one after another .

Its report released in 2021 shows that the RPA has seen a surge in its applications with a wider scenarios involved. While during 2018-2019, RPA was mainly used in the financial areas, it has reached out to more scenarios like human resource, IT, administrative management, etc during 2020-to 2021.

Data shows that 460 RPA robots were deployed during 2020-2021, up 110% in the overall RPA applications, saving 848,336 man-hours, with 1,708 man-hours each, compared to 1,335 man-hours during 2018-2019. And its ROI from RPA projects is also expanding.

The maturity of the RPA projects in the US federal government has improved quickly. During the 2020-2021 fiscal year, the federal government's RPA projects saw significant boosts in terms of RPA efficiency time saved, the number of projects, business process rebuilding, and security and regulatory compliance, especially in business efficiency, business process improvement, and  business process rebuilding, which demonstrated more notable performances..

In addition, more and more U.S. federal government agencies have set out to expand the RPA adoption, including the U.S. Department of Defense, Department of Agriculture, U.S. Army, Air Force, NASA, U.S. General Services Administration, and other important agencies which have already placed RPA in their practice scenarios, pushing RPA to be one of the most popular tools for their digital transformation.

Liu Chungang, CEO of ENCOO, believes that the success of overseas RPA companies has proved the value of RPA as a product to customers. More entrants are required to educate wide audiences about the value of RPA. We believe that this sector will be more mature with the influx of various players and healthy competition environment established.

02

Why is RPA favored by government agencies?

A report from Forrester Research says that 10% of government services will be automated by 2020. Globally, government institutions will expand the use of RPA to eliminate repetitive low-value tasks.

That may be the main reason why the US and Japanese governments have increased RPA adoptions.

In the United States, RPA helps save a lot of man-hours and human efforts in the federal government.

With the help of RPA, U.S. federal government agencies have saved 1.49 million man-hours in the 2020-2021 fiscal year, and RPA projects have increased by 250% compared with 2020, reaching 1,003.

The application of RPA in US federal agencies has achieved more than expected ROI (return on investment), as well as a stronger capability to handle government affairs, and end-to-end automation. With less manual work to be completed by federal employees, the workforce structure in the federal government has been reshaped, thus relieving the pressure in human resources.

Due to these advantages, the U.S. federal government is considering to further expand RPA applications by tapping into its deep potentials to broaden its coverage. Some key institutions have made it clear that they will continue to increase the number of RPA applications and scenarios to put it into more practices. They will  create intelligent digital workforce through attended or unattended RPA, to drive productivity and save time.

According to the federal RPA application maturity model, most institutions are reshaping their business process standards through RPA. At present, in order to better manage RPA, the institutions with RPA in place have begun to develop standard automated processes, evaluation criteria, and baselines for improvements, to push their RPA applications to a new level.

RPA has been put into place in dozens of business scenarios in Japanese government agencies and achieved satisfactory results.

According to the report released by Japan, RPA will be one of the most effective tools for Japan to solve the "2040 crisis".

By 2040 crisis, we mean that by 2040, the Japanese population over the age of 65 will reach a peak of more than 40 million, which means that there will be one out of three Japanese people that reaches 65 years old, and the labor force will drop to freezing point, or even negative growth.

Therefore, the Ministry of Internal Affairs and Communications of the Japanese government hopes that RPA can be widely used in various industries throughout Japan like physical robots to relieve the pressure on human resources.

As one of the earliest countries in the world to apply RPA, Japan has always been in a leading position in terms of investment, technology development, talent training, as well as application scenarios, scale, and maturity.

Japan is one of the countries with the lowest birth rate and the most aging population in the world. That is why it is actively using RPA to create Digital Workforce to make up for the shortage of labor.

03

What are the obstacles in RPA adoption?

The Forrester report also argues that not all RPA implementations in public sectors work as expected. Changes in management may occur in many organizations, which requires rethinking processes, standardizing tasks and reallocating work.

By studying the RPA applications in the US and Japanese governments, we can identify several common problems.

High cost in building and maintaining RPA applications. According to Japanese surveys, most RPA applications requires 1 million to 5 million Japanese yen (about RMB 54,000 to 270,000 yuan) to be built.

Similarly, their operating expenses amount to 1 million and 5 million Japanese yen per year. And most of the RPA projects require less than 1 million Japanese yen to be introduced and maintained.

Less professional talents to implement RPA projects. Lack of professional talents, insufficient financial budgets, and inability to accurately identify business processes fitting for RPA create obstacles for its implementation. Among them, the talent gap in RPA areas is the top concern.

Enhanced supervision over RPA required As the RPA is increasingly used, it requires a higher level of management in some sectors, especially in some special sectors where the applications of RPA should be strictly supervised.

For this purpose, security supervision mechanisms around data access, authentication, etc. should be established in every organization with the support of the IT department. At the same time, standard procedures should be formulated for data privacy management, data storage and access and use permissions, in order to protect data assets.

An effective linkage with IT required. In federal agencies in the United States, the RPA projects should be officially authorized to be applied in their IT environments. Even a single automated process created through RPA should also be approved by relevant departments to maximize the security and privacy of data.

So, the IT department plays a crucial role in pushing the RPA projects forward. An effective linkage should also be established with IT. Statistics show that 43% of RPA pilot projects have set up IT security approval mechanism.

Scenarios of RPA applications expanded through CoE (Center of Excellence) in most organizations. With the steady improvement in maturity of RPA applications, many federal agencies in the United States have begun to expand the automated scenarios through CoE, which can elevate efficiencies in RPA deployment, training and management.

In the face of such hindering factors as process approval, IT managers' hesitance, and so on, many organizations are looking into more Enterprise RPA solutions to run more of their business processes around the clock, and drive more efficient and user-friendly automation designs.

04

AI and Deep Learning: New Trends in RPA

Sun Linjun, CEO of Intelligence Indeed, believes that all industries are moving from information technology-driven to automation-enhanced, then to intelligence-powered. Currently, facts prove that the integration of RPA and AI can meet customers' needs at a lower cost, and bring immediate benefits.

RPA and AI are moving together.

With the deeper application of RPA, in order to meet the automation needs of more business scenarios, the US federal government is looking to integrate RPA with many AI technologies like as Natural Language Processing (NLP), Machine Learning (ML), image recognition, chatbots, etc. According to the survey, 32% of RPA projects will use Intelligent Automation (IA) solutions.

Another survey conducted by the Japanese government on a total of 1,788 government agencies also showed that, as of June 2021, 118 organizations had successfully introduced RPA, 152 introduced AI, 265 introduced both RPA and AI, and a total of 535 organizations had put RPA and AI into their practices.

As Gartner recommends, the next step should be the full integration of RPA with AI and ML to achieve Hyper Automation and reduce human intervention.

According to Gartner, the RPA market will see rapid growth over the next decade, while Hyper Automation will be an essential part of digital transformation.

Hyper Automation refers to the use of advanced technologies, such as AI, ML, RPA, etc., to complete almost all tedious and scalable tasks previously taken by humans, overcoming the weaknesses of single automated tools.

As such, Hyper Automation can be seen as a technology portfolio around the businesses, which can identify, review, and automate as many business and IT processes as possible, based on the coordinated use of multiple technologies and platforms, including RPA, low-code platforms, and process discovery tools, among others.

Hyper Automation cannot be the absolute substitute for human beings, Instead, it can set people free from repetitive, low-value tasks to focus on those of higher value to the organization.

It is often said that Hyper Automation is the successor of the traditional automation, and its market size is expected to reach $46.4 billion, growing at a CAGR of 21.7% between 2021 and 2031.

Deloitte also believes that the combination of automation technologies like RPA and AI will bring about Intelligent Automation (IA).

RPA also has limitations. For example, it can only follow processes based on  logical rules. It cannot identify the patterns in data, nor can it extract information from images, texts, or voices.

As shown in Deloitte's report, enterprises are now looking for next-generation solutions that leverage multiple advanced technologies and data science, such as AI, to make automation smarter and deliver more value to them.

Through Intelligent Automation (IA) technology, organizations can transform business processes to achieve greater speed and precision and automate prediction and decision-making based on structured and unstructured inputs to achieve higher productivity and lower costs, more targeted business operations, and better customer experience.

In the next 3 years, Intelligent Automation (IA) is expected to reduce costs by 22% on average while increasing revenue by 11%. Today, 47% of enterprises have incorporated the integration of RPA and AI into their Intelligent Automation strategies.

Some enterprises have also taken important steps on the road to the integration of RPA and AI.

Salesforce, the world's largest SaaS company, added Einstein Automate, an RPA tool, to its AI platform in 2021 to help automate processes, build workflows and connect data.

The new tool includes MuleSoftRPA, EinsteinDocument Reader, and DigitalProcess Automation. MulesoftRPA allows users to use robots to take repetitive tasks. Einstein Document Reader scans documents such as driver's licenses and I-9s. Users can take actions on the data with just a few clicks in Salesforce Flow. DigitalProcess Automation allows users to build branded digital experiences using drag-and-drop tools.

Appian started out as a Business Process Management (BPM) company and has expanded its business into low-code and RPA. As a famous low-code and RPA platform around the world, it has launched its enhanced AI document processing and developers' collaboration tools in its new build. Its low-code robotic process automation can be used to manage and monitor Appian bots as well as bots from third parties. Tasks can be automated using its low-code RPA and new libraries are available in AppianAppMarket. Intelligent Document Processing (IDP) handles unstructured data at scale, with optical character recognition built in its trained AI models.

Chen Yunwen, CEO of Data Grand, said that China's RPA market is still in its early stage on the whole. Only less than 1% of its intelligent RPA business market has been opened, while still 99% is yet to be exploited.

Inspired by the practices of RPA applications in the public sectors of the US and Japanese governments, we have expected a larger capacity in the RPA market with attractive financial gains, allowing for many more potential RPA players to tap into.