Robotic Process Automation (RPA) is the acronym used for the software “bots” that either reside on the user’s desktop (called attended robots); or in the background (unattended robots). There are 2 primary classes of work that fit the software robots well:
1) Attended robots work in conjunction with a human. Typically the process is started by a human to complete a task that is tedious, error-prone, and requires limited decisions. This robot will reside on the same machine as the user and effectively mimics the user’s keystroke steps at “machine speed”. We’ve discussed several symptoms and likely candidate use cases for when an attended robot will be most successful.
2) Unattended robots work in the background without human involvement. Think of large/bulk data file manipulation or batch data processing. These can be scheduled, or triggered on pre-determined events. These typically work from a designated input data file or system that contains structured or semi-structured data which can be added/edited/updated/validated in other systems automatically.
Today, let’s focus on the unattended robot usage patterns that are more common, and usually have a shorter path to ROI. Note that each pattern could be home for a collection of very specific vertical or horizontal use cases, these could be specific to an industry (Finance, Telecom, Healthcare, etc.) or even departments common to all verticals (HR, Sales, IT, etc.).
Unattended robots have parallels with some legacy technologies. We all remember the ETL, EAI, and SOA products for managing batch, bulk, or real-time data integration. Many of the “jobs” put in place years ago are still running and keeping the business going (whew!). These backend “jobs” are what unattended robots can do today in a fraction of the time and cost, with the newly added benefit of seamlessly crossing technology platforms/languages and providing Artificial Intelligence for processing less-structured data. These unattended robots can also take advantage of systems that do not have APIs or “typical” integration mechanisms since robots can intelligently read any UI screen, file, or even unstructured data (think eMail) across various technology platforms.
Regardless of industry vertical, horizontal, or business units, several “patterns” emerge that make excellent use of Unattended Robot capabilities. These usage patterns can be executed in the cloud or on premise, and can be stand-alone or even work with “hybrid” implementations where the unattended robot and attended robot interact together. These patterns extend beyond legacy use cases for data migration/integration since Artificial Intelligence (AI) components can be added to intelligently assist with data extraction, validations and cleanup from either structured or unstructured data sources.
There are great ways to extend the power of unattended robots since they work 24x7x365. As attended robots work mainly in-line with humans to support live business processes, the unattended robots work in the backend to support key data management operations. Some common unattended robot usage patterns are:
- System Monitoring – Unattended robots can be used to perform mundane human tasks by “reading” content and status in system consoles, logging into legacy platforms/system administration portals to monitor alerts, verify servers or databases are active, ensure APIs are responding, and even non-obvious things like verifying security cameras are working (picture exists) and performance monitoring by mimicking users’ login and navigation to ensure screens are rendered within a specified tolerance.
- Data Cleanup – Robots are good at repetitive work, so the data cleanup pattern supports use cases for examining tons of data to see if data is duplicated, mal-formatted, or even missing. This could be names, addresses, phone numbers, eMails, etc. Additionally, since robots are system independent, this pattern can be leveraged either within a single system, or across multiple disparate systems; and even extend to Office tools such as Excel, Word, PDF, or other non-standard data sources. Suspect data can be fixed/updated by the robot for low-risk scenarios, or saved to an ‘exception’ file for high-risk data so humans can review and approve before changes are made.
- Improve network/server loading – Unattended robots can work from a queue or even be scheduled to process data with minimal performance impacts to the production servers or network. With more remote workers than ever before, networks are stressed under the real-time performance demands. System/data updates can occur offline and sequenced after users are “done with their day”; this helps to defer or even eliminate expensive network upgrades and take advantage of lower VM costs for after-hours processing. This pattern supports scenarios for after-call work in many verticals where humans can hand-over the necessary system updates to a robot for batch processing.
- Manage Sensitive Data – Many industries or at least departments work with data that involves PII or other highly regulated (e.g. HIPPA) content that may create corporate risk when humans have direct access. Unattended robots can process such data without risk of exposure or unintended modifications. Precision, security, and speed are all core robot attributes that strongly support data privacy and ensure security protocols are followed. This is very beneficial in RPA for Human Resources and RPA for Finance / Accounting scenarios.
Unattended robots can work the “other 16 hours” of the day, outside the time of typical human interactions. That said, during typical “working 8-hours” the unattended robots can also interact with attended bots or human “trigger events” that can help support front-office process as well. Beyond attended and unattended, hybrid robot patterns exist that mix the humans, attended, and unattended into a common workflow. Some companies view the unattended robots as a path to avoid (or delay) the purchase of attended robots where data processing can occur before or after the “required” human interaction. Clearly some business processes take human empathy, knowledge, decisions, or approval – so unattended robots can perform the mundane pre-cursor prep work or post-human (e.g. after call) processing and system update work.
In summary – all robots, whether attended or unattended; excel when the process is repeatable, decisions are rules-based, and predictable data structures exist - these will be our first list of patterns to consider for automation. The “Software” Robotic Process Automation (RPA) platforms are constantly adding features and extending capabilities to make the robots smarter (AI, Machine Learning, Natural Language Processing, etc.). At Optezo, we want to remove the technical complexity and pricing confusion by offering and bundling the RPA as a service (RPAaaS). This simplification allows our customers to focus on identifying the key patterns and leveraging the Optezo process catalog for picking the optimal use cases.
Engage our experts to define, design, build, and manage the support and robot infrastructure so you can focus on your business. Most businesses can benefit from leveraging the RPA patterns we’ve outlined above, with significant value related to savings or even cost avoidance while improving accuracy. Take a look at how we approach RPA differently.
