Robotic Process Automation (RPA) is intelligent automation software that works with existing applications to perform the same tasks that are done by humans today. RPA uses software 'robots' to learn and execute a variety of skills ranging from repetitive, high-volume tasks to complex ones involving data transformation and logic.
Once implemented, the robots become a virtual workforce controlled by the business operations teams.
RPA was designed to work with a variety of technologies and bridge the gap between systems and processes. Leveraging advanced tools like Optical Character Recognition (OCR) and artificial intelligence, RPA has the ability work with just about any application (web or desktop). Combined with integrations to databases, file systems and email, the RPA platform is able to support key processes end-to-end
RPA stands for Robotics Process Automation. RPA and its robots (or bots) is a technology that can be configured to “observe” the way a trained user performs a particular task and the various decision points involved in accomplishing that task, and to replicate the process. See our article on RPA basics, where we offered a high level description of what RPA is (and we’ll take deeper dives in upcoming articles).
RPA is ideally suited for tasks involving transactional data from a variety of sources.
Some examples of its current application include: Invoice processing, accounts payable, travel and expenses, claims processing, payroll input and change of (address, address, name, etc.). It is estimated that up to 45 percent of the activities companies pay people to perform can be automated through RPA.
Businesses are best matched with RPA if employees are consumed with high volume repetitive tasks that follow structured business rules that generally result in different outcomes. These are good guidelines: You are able to describe the work. This doesn’t mean your documentation exists or is current. The task could be described by recording a user performing their work on a computer including how they handle exceptions. The work is rules-based rather than subjective. Robots need to be prepared (aka, taught, trained, configured) to perform specific actions on your systems. While getting closer each year, current technology is insufficient for a robot to determine on its own what to when faced with a new situation.
The work is performed electronically. It doesn’t matter how many different applications are required or whether they are in-house, cloud-based, Citrix, desktop or mainframe.
The required data is structured (could be structured). If not, you may be able to utilize an OCR and/or cognitive application capable of structuring the file. Alternatively, you could have people enter the data into a structured format.
It is not necessary for IT teams to change any existing systems for an RPA solution. RPA software can non-invasively interact with target applications through same user interface utilized by humans. Leading RPA software tools can interact with most major mainstream and legacy thin and thick client applications based in Java, Mainframe, Web, and Windows. If the robots were run off-premises, IT team would need to grant robots with system access as they would do with a new employee and keep the underlying applications running.
Leading automation solutions have native multi-level, configurable, auditable logs in addition to reporting and alerting features. These allow organizations to meet or exceed their industry-specific regulatory requirements including SOX, HIPAA, and PCI DSS.
Leading RPA tools allow for out-of-the-box and customizable role-based access rights that support separation of duties. Administrators, developers, and other end users are restricted based on their permissions.
Leading RPA tools use server platforms with application level encryption that is Federal Information Processing Standard (FIPS) compliant including AES 256.