什么是超自动化?

Jeff Erickson | 技术内容策略专家 | 2023 年 5 月 5 日

For years now, businesses have been collecting data on their daily operations, measuring manufacturing lines, healthcare interactions, shipping logistics, financial transactions, and countless other critical details. There has long been a market for business process software that uses that data to better understand those operations and simplify, standardize, and automate where possible. Now there’s a new opportunity: Take all this valuable operational data and business process expertise and use it to train artificial intelligence and use AI-backed applications to better anticipate and then react to business ups and downs. Doing so successfully could deliver a powerfully disruptive business advantage, which analysts at Gartner have dubbed hyperautomation.

What Is Hyperautomation?

Hyperautomation is a business technique for wringing more speed, efficiency, and accuracy out of daily work operations. It is related to another trend: intelligent automation, which is the technical process of combining robotic automations with artificial intelligence (AI) and machine learning (ML) to mimic human interactions and automate complex processes. Hyperautomation is the business discipline built around intelligent automation, helping organizations tap into their storehouse of operational data to identify and automate more business and IT processes.

Unlike typical automation, hyperautomation gives businesspeople a way to access operational data, integrate data sources, and use AI services to automate complex, nuanced business process. These can include customer service interactions, document routing, shipping logistics, business analytics, and many other processes. With data-backed capabilities, such as artificial intelligence, natural language generation, computer vision, and anomaly detection, hyperautomation seeks to orchestrate a business process that requires interpreting human language, advising best options, and even analyzing a series of steps and incorporating bots to automate them. The goal is to deliver the best business outcomes with the least cost while improving the accuracy and speed of business operations.

Hyperautomation overview, description below
The four key ingredients in hyperautomation are people with expertise in the task to be automated, operational data from business applications, automation tools, and artificial intelligence.

Hyperautomation

  1. Expert people: Industry knowledge and task expertise
  2. Business applications: Operational data and data integrations
  3. Automation tools: Data orchestration and robotic process automation
  4. Intelligence: AI and ML to mimic human expertise

Key Takeaways

  • The goal of hyperautomation is to improve the efficiency, accuracy, and speed in an organization’s daily operations, including adding real-time feedback where viable to help an organization respond to or even anticipate business change.
  • Hyperautomation is different from and more capable than simpler robotic process automation (RPA).
  • Hyperautomation combines business process automation, software integrations, and large stores of operational data to train artificial intelligence and automate complex business processes.
  • Hyperautomation can bring new levels of efficiency and accuracy to tasks such as handling customer service inquiries or intelligently routing goods or documents in a wide range of business operations.
  • Hyperautomation requires thorough planning and ongoing assessment and analysis to refine outcomes and achieve the desired return on investment.

Hyperautomation Explained

Organizations use hyperautomation to squeeze maximum efficiency from daily operations and get the best results. Although its goal is simplicity for the business user, hyperautomation involves a complex orchestration of multiple technologies, tools, and platforms. According to analysts at Gartner who coined the term, hyperautomation involves: artificial intelligence (AI), machine learning (ML), event-driven software architecture, robotic process automation (RPA), business process management (BPM) and intelligent business process management suites (iBPMS), integration platform as a service (iPaaS), low-code/no-code tools, packaged software, and other types of decision, process, and task automation tools.

A hyperautomation practice is a way for a business to get more value from its industry expertise and the mass of operating data it collects every day. By integrating data flows and training AI, businesses can increase efficiency in their day-to-day operations and support more effective interactions between employees, customers, and partners.

As you can see, hyperautomation requires combining many disciplines and deep expertise in the operation being automated. The result, however, can provide a new level of efficiency in customer and partner interactions that can lead to cost savings and competitive advantages.

How Does Hyperautomation Work?

A hyperautomation project involves identifying workflows that would benefit from automation, sourcing, and integrating the right operational data, choosing the appropriate automation tools, reusing proven automations where possible, and extending their capabilities with various forms of AI and machine learning, such as anomaly detection, computer vision, or natural language processing.

The goal of hyperautomation is to mimic the way clients or employees would interact with applications and blend into the process. For example, by recording how people perform a task, a bot can be created that automates that action; with AI, that action can include understanding intent in a client’s natural language and making decision on next steps in a workflow. Over time, the data from these digitized operations can be analyzed to find hidden opportunities for improvement in the business process. As tasks get more complex and the speed and accuracy of business operations improves, you’ve moved from automation to hyperautomation.

OCI delivers high business value

According to IDC, OCI can provide a 474% five-year ROI and a 53% reduction in TCO.

Hyperautomation vs. Robotic Process Automation (RPA)

Robotic process automation (RPA) is a system that allows people in an organization to create computer bots to replace human interactions and take away mundane, repetitive tasks in their work. If someone is regularly copying and pasting text or constantly moving documents from one file to another, a computer bot can be generated to take on those steps while the employee moves on to other tasks.

Hyperautomation is, effectively, the orchestration and enhancement of RPA to accomplish broader and more complicated goals. To automate tasks and later build more complex hyperautomations, the IT team provides standardized repositories of operational data and APIs for integrating data from multiple sources. Businesspeople get a low-code or no-code platform to drag and drop data and integrations to build an automated workflow. Organizations often govern the process with an automation center of excellence that vets ideas, coaches the participants, and provides support.

Hyperautomation adds a layer of artificial intelligence that is trained on and informed by the large volume of historical and near real-time operational data. The use of AI enables automations to interact with customers, partners, and employees in ways that can, for example, understand intent, quickly source accurate information, take the appropriate response, and communicate in natural language.

Why Is Hyperautomation Important?

Hyperautomation is an application of sophisticated AI that has the potential to revolutionize business. It opens opportunities to give businesses a competitive advantage by offering new levels of efficiency. Hyperautomation helps companies make better use of all the operating data they collect and store; it lets them take a smarter action in the moment by using event-driven software, rather than just using data to look back and analyze the past. For example, a port can track and move containers faster and more accurately by using computer vision to automatically identify and measure a container as it enters a port. Or an insurance firm can expedite a claim using intelligent character recognition to look at the text of a document and then automate a document flow, flagging only a small number for review by an employee. Likewise, finance, healthcare, manufacturing, and online retail can all be made more efficient by enabling faster, more accurate interactions with customers, patients, and suppliers by using business automations that also reach into a supply chain to foresee needs, order goods, fill out documents, and suggest next steps to clients or employees. In all these areas, hyperautomation is a competitive advantage that reduces the burden of repetitive tasks, lowers cost, improves accuracy, and leads to innovations.

Benefits of Hyperautomation

As hyperautomation takes hold in businesses, they’re seeing benefits in numerous areas.

  • Business speed: Improves productivity and efficiency across an organization by automating ever more complex operations.
  • Improved accuracy: Helps organizations process large volumes of data quickly and accurately without missing details or introducing errors into their systems.
  • Reduced reliance on manual labor: Lowers costs and frees employees from tedious tasks—whether in the field or the office—so they can concentrate on more important projects.
  • Additional use of data and IT infrastructure: Leverages existing operating data and IT infrastructure to achieve better business results at lower costs with greater speed and accuracy.
  • Process improvement: Can provide more consistent and predictable actions in a business process and use AI to analyze daily operations and make swift business decisions, as well as process improvements over time.
  • Improved customer service: Pleases customers and clients by understanding intent, sourcing information quickly and accurately, and communicating with natural human language. Having people ready to jump in for complex or emotional situations still matters, but many situations are deftly handled by hyperautomation chatbots.

Challenges of Hyperautomation

Hyperautomation promises much, but it takes thorough planning and a commitment to data management to make sure the right data is used to drive responses. Without these, it can become a burden instead of an asset. Potential challenges include:

  • Costly business process re-engineering: Bringing a business process into line so it can be automated requires time and planning, as well as buy-in from employees. That could mean a lengthy change management strategy with regular and consistent communications to effected workers.
  • Technology upgrades: To move from simple bot automations to hyperautomation requires investment in new technology such as event-driven software, an effective integration platform to bring the right data together, and low-code development tools. It also requires expertise in business process automation and artificial intelligence. For example, the ability to do real-time analytics is vital to operating and judging the effectiveness of the automated functions.
  • Required analytics: Part of the planning for hyperautomation is an upgrade in data management and data analytics, which will give you the information to measure effectiveness and check for compliance throughout your automation process.
  • Required security infrastructure: Hyperautomation touches many different systems in an organization, so it needs to be tested to guarantee that any bots or systems connected to the operation can’t be infiltrated by malicious actors.
  • Bias: When applying business process automation to make decisions—be they about loans, hires, purchases, or insurance rates—great care must be taken to ensure the algorithms don’t project unintended biases.

Hyperautomation Use Cases

Across industries, hyperautomation is proving it’s worth by helping organizations cut costs, improve service levels, and lower risks. Here are five real-world examples:

  • Retail: After experiencing triple-digit growth, Brazil’s Facily found that customer service had become a challenge. The social commerce marketplace built automations that integrated financial management, inventory, and logistics, and powered the core functions of its online marketplace—creating a fully auditable trail, from ordering and approvals to accounts payable, leading to faster order fulfillment and better inventory control and purchasing.
  • Finance operations: Lyft used automation to cut—by more than half—the time it takes to close its books on the revenue component of its ledger. The company expects to cut the time for its full financial close in half and then in half again as it continues to improve its processes.
  • Supply chain: Because it insists on fresh ingredients, Chipotle requires a responsive supply chain. They used automation in two ways: First, automations track and forecast needs constantly and accurately, allowing suppliers to trust the forecasts and plan accordingly. Then they automate order tracking for takeout and delivery so they can reconcile, without human verification, every online transaction with delivery partners—saving millions of dollars a year in reconciliation activities with drivers.
  • Healthcare: Danish Healthcare provider Coloplast uses marketing automation to communicate with and support ostomy and catheter users. The company achieves engagement levels well above healthcare industry averages. Their personalized, interactive service program is designed to help improve users’ conditions and quality of life with accurate and timely information through automated communications via multiple communication channels, including email, web, direct mail, and phone.
  • Customer service: Automation lets Razer give tech-savvy gamers the self-service tech support they prefer. The majority of Razer’s customers said chat was the most convenient way to contact the company, but the company’s most mature communication channels were phone and email. Razer brought in chatbots and AI to optimize and automate its service process. Now the company services 50% of customer inquiries with automated chat while intelligently routing the rest to other communication channels.
  • Data security: Adenza’s AxiomSL business unit is a global provider of data and risk analytics software for regulatory reporting in the financial services industry. They use machine learning automations in their database infrastructure to eliminate most database maintenance activities and provide automated scaling and tuning as well as an automated process that encrypts sensitive and regulated data, patches databases for security vulnerabilities, and helps prevent unauthorized access.

Starting the Hyperautomation Process

For a business process to go from manual to hyperautomated takes the commitment of many people, as well as a whole lot of data and other technology. Here’s a breakdown of the high-level steps involved in hyperautomation.

  1. Establish your automation needs and opportunities.
    Hyperautomation starts with your established business processes. Gather information on your current operations and business rules and how they compare to industry best practices. Interview people on the front lines to learn where the bottlenecks are or where automation can alleviate repetitive processes. Find the simple robotic process automations that might already exist. This information will help you plan the hyperautomations that will serve your organization best.
  2. Gather your data.
    Much of hyperautomation is a process of integrating data sources and creating real-time data flows. This will help you create the handoffs and approval chains in complex automations. Use your stored operational data to train ML models.
  3. Identify your tools.
    Building automations will include business process automation services, data integration tools, low-code development platforms, ML training platforms, analytics tools, and in some cases, Internet of Things (IoT) sensors. These will be tied into your business application data to automate interactions between and within those applications.
  4. Organize your personnel.
    Identify project leads in different areas of the business as well as the leaders who will sign off on the automated approval processes. These people help you identify and organize processes to automate and will be most interested in the analysis of efficiency gains in their areas. Note that automations can leave front line workers confused about their roles. Help them establish new routines that take advantage of automated processes.
  5. Implement, measure, iterate.
    Hyperautomation is a collaboration between human expertise and machine learning algorithms enabled by real-time data flow and data analytics. Keep an open line of communication with the frontline workers to measure the effectiveness of any automations and be prepared to iterate. In time, you will be able to rely on data analytics to show just how the automated processes are helping the business to be both more efficient and more agile.

6 steps to hyperautomation diagram, details below
Hyperautomation is the end result of many steps. These steps include starting small with simple task automations and then designing more complex automations with event processing. Lastly, train AI and ML algorithms to take on more tasks and continue to iterate to improve outcomes.
Illustration of stairs ascending from lower left to upper right. On the lower left step is a person standing with hands on hips. On upper right step is a person standing waving a flag in triumph.
  1. 1: Start small. Automate simple tasks according to business rules
  2. 2: Think big. Use business process automation to envision new automations
  3. 3: Use data. Integrate data sources into process flows. Train AI and ML algorithms
  4. 4: Create the path. Add event processing architectures to orchestrate automations
  5. 5: Apply intelligence. Apply AI and ML algorithms within established workflows
  6. 6: Iterate. Improve algorithms with more data

Get the Benefits of Hyperautomation with Oracle

When it’s time to bring hyperautomation to your organization, you’ll want trusted tools for process automation, IoT, data management, and AI services. A good place to start your effort is Oracle Cloud Infrastructure Process Automation, which helps developers and business experts automate approval workflows that span ERP, HCM, and CX systems. To pull in all the data you need to fuel this hyperautomation, you’ll need an integration service, such as Oracle Cloud Infrastructure integration services, that’s capable of connecting any application or data source. Regardless of your industry or use cases, OCI gives you the tools needed to simplify repetitive tasks with reusable business rules, prebuilt integrations, and low-code designs.

Hyperautomation FAQs

Why implement hyperautomation?

Hyperautomation is a business strategy that delivers new value from operational data. It combines process automation and data integration expertise with AI and ML capabilities to achieve greater speed, efficiency, and accuracy in daily work operations. It does this by automating complex workflows, such as document management, customer service interactions, and many other processes to deliver a competitive advantage.

How is hyperautomation achieved?

Hyperautomation involves orchestration of multiple technologies, tools, or platforms. It combines business process automation platforms with technologies such as robotic process automation (RPA) and advanced AI and ML technologies.

How can I get started with hyperautomation?

Hyperautomation is an extension of business process engineering. Look for a partner who knows process engineering and offers the AI and ML services and integration tools you’ll need to move from automating tasks to hyperautomating whole business processes.

What are some best practices for hyperautomation?

Best practices for establishing hyperautomation in your organization include identifying workflows that might be automated, choosing the appropriate automation tools, reusing proven automations where possible, and extending their capabilities with various forms of AI and machine learning. You’ll also want a feedback loop to verify that automations are achieving their goals and are improved over time.

Delivering on the Promise of Modern Cloud Economics

Learn more about how OCI's approach creates a differentiated cloud experience and delivers on the promise of modern cloud economics.

注:为免疑义,本网页所用以下术语专指以下含义:

  1. Oracle专指Oracle境外公司而非甲骨文中国。
  2. 相关Cloud或云术语均指代Oracle境外公司提供的云技术或其解决方案。