Patrick Sullivan, Aptive’s VP of Digital and Technology Solutions, and his award-winning team work with federal government clients to collaboratively build solutions, enable disruptive engineering through emerging technology, and harness data to deliver value. In this exclusive guest blog, Patrick discusses the need for modernization across the public sector, our partnership and work with a large cabinet level federal civilian agency, and predictions for the future of digital innovation within the public sector.
Why Intelligent Automation is a Must for Federal Agencies
Most government activities are driven by forms. Need to file your taxes? Fill out a form. Get a passport? Fill out a form. Access your GI Bill benefits? You guessed it… fill out a form! Every day, agencies across the federal government must efficiently and effectively process millions of applications, claims forms, tax documents, and other pages. These come in a wide variety of formats with messy handwritten text or low image quality and require a large data entry lift.
Through our partnership with Hyperscience, our automation and analytics toolbox now includes modern, best-in-class intelligent document processing. These capabilities can automate over 90% of document processing and data extraction without sacrificing accuracy. With intelligent automation, agencies are able to decrease processing times and reengineer their workflows to be more efficient and effective. The result is freeing up valuable employee time to focus on more strategic, innovative initiatives that better serve constituents.
Further, many government agencies have digitized historical records but may not be doing anything with that data. By using intelligent automation, these agencies can “unlock” and extract actionable data for use in a way that enhances their services, like detecting fraud, waste and abuse, or establishing predictive analytics.
On Our Work with a Large Cabinet Level Federal Civilian Agency
We originally partnered with Hyperscience to deliver Artificial Intelligence (AI) and Machine Learning (ML)-based automation to a large cabinet level federal civilian agency.
We were tasked with extracting data from handwritten forms and paperwork to help the agency achieve its automation goals to reduce the benefit claims backlog and speed up the time it takes to adjudicate claims. We explored several Optical Character Recognition (OCR) products that didn’t quite fulfill our needs. Then we tested Hyperscience.
We saw great results from our first prototype, and we knew this was the tool – and the partnership – that would help us solve their mail intake challenge. What stood out? The true incorporation of AI/ML to recognize forms and extract data with a high degree of accuracy. It’s made a world of difference in the outcomes the agency’s been able to achieve for veterans.
The Hyperscience platform allows the agency to:
- Configure accuracy thresholds
- Improve performance over time
- Consult a comprehensive analytics dashboard for a unified view of human and machine performance
These scalable, user-friendly features contributed to reducing the time it takes the agency to sort a claim from 10 days to less than half a day, which is a major win as the agency receives roughly 50,000 compensation and benefits-related mail packets each day.
Today, the platform processes more than 1,500,000 documents monthly, extracts over 80 million fields, and straight-through processed 2 million forms in the first year.
The Future of Public Sector Innovation
The early government adopters of modern and emerging technology now have evidence-based results and are inspiring others to follow suit. We’re seeing more and more technology pilot projects. These allow testing to be done on a smaller scale, reduce risk, and provide data-driven results that can be shared with leaders to inform their decision-making related to larger-scale investments in cutting-edge technology.
I’m excited to see more agencies and programs realize the benefits of using AI and ML to achieve their missions. We’re finally beginning to see a focus on data standardization, data integration, and Cloud migration and computing. When you have true integration happening (rather than data silos), you’re looking at the foundation for building complex AI/ML-based capabilities that can use an organization’s data to automate repeatable tasks, augment the work performed by its employees, and help make predictions. By using data to identify trends and patterns, you can provide better services to constituents and other stakeholders.
Patrick Sullivan is the VP for Digital and Technology Solutions at Aptive Resources, helping the company chart a course to revolutionize the future of federal healthcare and other important services for veterans. In his role, he focuses on enhancing systems and modernizing service delivery models through digital transformation. Connect with him on LinkedIn.