By Des Nnochiri
Robotic process automation, or RPA, promises to increase efficiency and improve work rates at reduced cost to the enterprise. In this article, we’ve assembled eight of the top RPA tools currently on the market. Of course, there are considerations to bear in mind before implementing this emerging technology.
What Is Robotic Process Automation?
Robotic process automation (RPA) is an attempt at streamlining enterprise operations and reducing costs by automating mundane or repetitive business tasks through the application of technology. This technology input is governed by business logic and the structured use of data. Use cases range from small-scale (e.g., configuring automated responses to certain types of email) to larger-scale deployments involving thousands of bots (e.g. automating the features of an enterprise resource management or ERP system).
Initially, the financial services sector led the way in adopting RPA technology. The tools are now used across a wide range of industries, including retail, human resources, and health care.
RPA tools enable organizations to configure software or “bots” that capture and interpret applications for transaction processing, data manipulation, communications with other digital systems, or the triggering of automated responses. These bots typically require no custom software or deep systems integration and are therefore low-cost and easy to deploy.
As part of a larger ecosystem known as intelligent automation, or IA, RPA tools may be used in conjunction with cognitive technologies such as speech recognition, natural language processing, or machine learning. Implementations like this can allow tasks that would in the past have relied on human judgment and intervention to be partly- or fully-automated.
RPA tools tend to be “trained” rather than programmed. This training comprises a learning process where, by observing how human operators interact with a system, the software comes to associate certain actions with a logical set of clicks. Functionality like this enables organizations to reduce labor costs and human error. David Schatsky, a managing director at Deloitte LP, cites the case of a bank which, by adopting RPA technology late in 2018, added a transaction-handling capacity equivalent to more than 200 full-time employees at approximately 30% of the cost of recruiting more staff.
According to figures from Gartner, Inc., the RPA market will top $1 billion by 2020, with robotic process automation and artificial intelligence helping to reduce employee requirements in business-shared service centers by 65%. Adoption is increasing as organizations realize that robotic process automation can not only streamline business workflows through the elimination of tedious manual tasks, but that it can do so without requiring them to completely re-engineer their legacy systems. Safeguards for troubleshooting and data security may also be worked into the robotic software layer.
For enterprises looking to capitalize on the use of RPA technology, there are several options currently available. These include the following:
1. Automation Anywhere
(Image source: Automation Anywhere)
Many of the tools offered by the Bot Store at Automation Anywhere rely on third-party libraries such as Microsoft Azure’s image analysis API. The store offers a large selection, including tools for extracting information from spreadsheets, files, and web pages. Automation Anywhere bots are also available to store this information in databases for issue tracking, invoice processing, and other business processes.
The Community Edition of Automation Anywhere is free for small businesses with a limited workflow and offers a testing ground and introduction to RPA technology.
(Image source: AutomationEdge)
AutomationEdge offers a selection of over 400 ready-made bots, focusing on databases such as SAP, Excel, and live chat applications. Intelligent Automation at Automation Edge incorporates AI, machine learning, ready bots, and chat bots, which enable faster speeds and up to 80% reduction in cost with almost no errors. A visible return on investment (ROI) typically manifests after four weeks of deployment. Many of the available bots are pre-configured for specific industries or business units such as customer relations or human resources.
For those who want to test the functionality, AutomationEdge offers the F3 RPA Bot, a free version of the platform limited to a single bot.
(Image source: Datamatics)
Bots on the Datamatics platform originate with TruBot Designer, a software tool enabling anybody to design and create bots. It incorporates a visual recorder, allowing business users without any programming knowledge to record the steps in a particular process, then click to create the bot. The Smart Library in TruBot Designer comes with pre-defined drag and drop components.
Business users can also collaborate with technical developers on the platform to design more complex bots with system-generated code fine-tuned to suit their business requirements. Advanced debugging tools and development options, such as the Integrated Development Environment (IDE) built on Microsoft Visual Studio, are available for programming complex processes.
The actions of all the bots created can be coordinated with the TruBot Cockpit, which deploys the tools and monitors their actions.
(Image source: IBM)
IBM’s range of options for automating business tasks takes the form of a set of separate products. Information input occurs via IBM Data Capture, while the flow of data is governed by Business Automation Workflow. IBM Operational Decision Manager (ODM) is an advanced business rules management system that helps users discover, capture, automate, and configure frequent, repeatable business decisions. The software adds real-time decision making to day-to-day business operations.
IBM Operational Decision Manager on Cloud is the Software-as-a-Service (SaaS) offering of the IBM suite of ODM solutions. The platform can host multiple environments and lets users customize the instance to match their staging process. Additional execution environments such as a user acceptance test or a pre-production environment are also available. The cloud offering comes as a monthly subscription package, offering a number of ready-to-use solutions as standard.
(Image source: Kofax)
The bots of the Kofax platform from ImageTech Systems can extract and act on data from many of the standard file types, including Excel, JSON, CSV, or email. Bots can be spun off into smaller tools called Kapow Kapplets for the local handling of focused chores. With Kofax, it’s possible to automate virtually any activity that involves accessing and acting on information.
Using the Robotic Synthetic API, code written in Java, Python, or other programming languages can reach out to the bots and give them instructions, allowing greater interaction with existing network stacks. Standard analytics and a reporting dashboard are available for monitoring bot behavior and fine-tuning.
(Image source: Nintex)
The Nintex collection of RPA bots includes tight integration with Office365, Salesforce, and Adobe tools, giving the platform a heavy emphasis on document and signature handling. Besides working with what Nintex describes as “logic-driven documents,” it’s also possible to automate standard data sources and use Nintex Hawkeye™ process intelligence and third-party services to gain insight into the effectiveness of your processes.
The platform has an impressive track record. In The Forrester Wave™: Digital Process Automation for Wide Deployments, Q1 2019, Nintex received the highest scores possible in 5 top-level criteria including vision and strategy, partners, commercial model, revenue generated by DPA wide sales, and number of enterprise customers. Nintex was recognized as a leader and ranked highest for strategy out of all 10 vendors evaluated against 33 criteria.
(Image source: UiPath)
The open and extensible UiPath platform features hundreds of built-in, customizable, shareable activities, and deep integrations with ERP, BPM, and AI technologies. Bots of the UiPath ecosystem are designed to be installed on Windows desktops and servers, where they can assume any of the tasks that might normally be done with the screen. A process known as the Orchestrator triggers them in response to pre-configured events.
With machine vision tools that can extract information from images or screen shots, UiPath features built-in intelligent optical character recognition (OCR), integrated NLP services from Google, Watson, and Microsoft, with deep learning to reduce maintenance. RBAC, encryption everywhere, and Veracode-certified code provide security and auditing.
(Image source: WorkFusion)
WorkFusion bots learn their tasks on Windows machines, then go to work on a Windows server using a mixture of repetition, OCR, and some artificial intelligence. The platform architecture has already been configured to meet some common challenges like anti-money laundering, insurance claims, coverage decisions, and banking services.
To give new users a taste for how the environment operates, WorkFusion offers the RPA Express Starter, a free tool limited to one bot running locally and with no access to the more sophisticated machine learning bots on the platform.
Issues to Consider
It should be noted that, as with other forms of robotics, RPA technology has the potential to eliminate jobs and present human resource managers with difficult decisions concerning their deployment of talent. Forrester Research estimates that as of 2019, RPA software will threaten the livelihood of 230 million or more knowledge workers, or approximately 9% of the global workforce.
Bot implementation has also proven to be more complex than many organizations had initially hoped, particularly where large numbers of bots—in excess of 50 for a single enterprise—are concerned. In addition, there’s not necessarily a direct correlation between the percentage of business tasks automated and the savings in cost for the business.
Change management is also a concern, with poor communication of the purpose and scope of bot deployment for the enterprise having as negative an impact as poor design of the bots themselves.
That being said, robotic process automation is evolving. Bots are becoming smarter and putting less burden on the user to train them properly. Artificial intelligence routines are also being employed to help look for patterns that will make the bots and RPA tools of the future faster and friendlier.