Page 26 - FCW, May 2017
P. 26

GUEST COLUMN
SYSTEMS DEVELOPMENT
Build a Case for Intelligent Automation
Intelligent automation, at several levels, can help agencies streamline operations and reduce costs.
Kirke Everson
Government Intelligent Automation Lead
unstructured data. This is where natural language processing and learning capabilities or “learning assist” come into play.
These are different from the quick hit tools of RPA. These technologies typically require integration
with the organization’s existing
IT infrastructure. They have the potential to transform the back office operations and automate the learning process itself. Think about a chat
bot on a web site or mobile device that helps a citizen engage an agency through text or voice chat. It can
also automate routine tasks such as password resets.
The third class of technologies is reasoning cognitive automation, the most advanced class of intelligent auto- mation. While still the least mature, it also offers the greatest potential. This cognitive solution has the ability to learn and solve problems using artifi- cial intelligence, machine learning and natural language processing. These technologies ingest massive amounts of data to formulate hypotheses, which would be too complex or time consuming for a human.
IBM Watson, which can analyze all forms of data, continuously learns, and gains knowledge over time, is
a good example of this advanced technology. Reasoning cognitive automation can significantly help with decision making and enhance mission delivery. It would be like having an expert by your side to
help make challenging decisions
SPONSORED REPORT
KPMG LLP
Agencies are having to do more with less, but still
must deliver valuable and efficient services to meet the rising expectations of the citizens they serve. It’s a familiar trend. This environment makes a compelling case for intelligent automation, which can help agencies improve interactions with their constituents, enhance employee job satisfaction, and provide accurate and prompt responses to citizen queries.
Intelligent automation is a spectrum of technologies ranging from software bots to automate routine, repetitive tasks, such as data entry into multiple systems; to complex, data heavy, cog- nitive processes, such as IBM Watson. Many agencies have begun using these tools in areas such as call center operations and in the back office.
With shrinking budgets, a smaller civilian workforce, and increased de- mands for better services, they really can’t afford to do otherwise. And with the growing support and economic viability for cloud computing, and greater availability of APIs means de- ploying intelligent automation is more affordable and scalable for agencies than it has been in the past.
CLASSES OF INTELLIGENT AUTOMATION
There are three classes of intelligent automation technology, ranging from simple to complex. Agencies are already using facets of each to varying degrees.
The first class is known as robotic process automation (RPA). This leverages technologies or software bots to automate routine, repetitive tasks and processes such as when a person cuts and pastes data from one form or system to another.
These tools leverage capabilities such as workflow, OCR, rules
engines and event data collection to automate existing manual processes by mimicking actions of a user. This means it can work with existing IT architecture without compromising it. This technology is a good entry point into using intelligent automation. Agencies can then move to more widespread use from there.
The second class is learning cognitive automation. This includes tools and technologies to deal with processes that may involve a high number of complex transactions and require a deeper level of analytics involving structured and


































































































   24   25   26   27   28