‘SecDevOps’ and RPA Drive IT Modernization Efforts at DHS

Despite potential for pandemic-induced disruptions, the Department of Homeland Security’s IT modernization strategy accelerated in 2020. The agency already moved about 75% of its network traffic to a new enterprise cloud-based network, and it hopes to finish the transition from legacy IT systems to the new network by early spring, said Deputy CIO (and current acting CIO) Beth Cappello.

“We have to ensure our mission capability is available from wherever our mission is operating,” Cappello said at an FCW Pillars of Modernization event last week. “Continuous improvement, modernizing this portfolio while supporting our workforce through the pandemic is incredibly important, and it also means we have to ensure we are securing our environment and modernization efforts.”

Capello said she expects the new enterprise services contract to “provide significant cost capability trade-offs and allow us to expand our core transport.”

“This network modernization really is key to our cloud application infrastructure and moving away from that aging physical infrastructure,” she added. “We’ve been able to reduce our physical hosting environment and increasingly leverage infrastructure for DHS applications. Accelerating cloud adoption remains a key part of our modernization strategy. It drives key outcomes, especially during this time of remote work. Remote access and scalability have helped us continue our mission.”

Capello said “SecDevOps” allowed DHS to maintain a viable telework environment while progressing modernization efforts.

“At DHS, security is first and foremost,” she said in reference to SecDevOps. “When we talk about security, we’re focused on modernizing and transforming our tech assets, and a key component of that is developing forward-looking architectures that will lead our IT investments into the future. We believe delivery-actionable guidance … will be able to expedite functionality out of the mission and reduce the complexity of the solutions our operators need most. It’s a major cultural transformation necessary to prove our reliability, agility and resilience. It’s an extension of our robust Agile development structure in place throughout the department.”

At the Federal Emergency Management Agency, robotic process automation and machine learning are driving IT modernization efforts.

“FEMA’s mission is helping people before, during and after disasters,” said Shang-Jeo Gaublomme, Financial Systems Modernization Branch Chief at FEMA. “It’s entirely integrated with response, recovery and resilience teams to distribute funding quickly and precisely for any disaster. In addition, we also need to safeguard public funds and prevent fraud.”

During the FCW Pillars of Modernization event, Gaublomme described time-intensive processes — like aggregating data, reconciling data and manual data entry — bogging down FEMA’s productivity. A bot, she said, frees up hours of time for FEMA employees to focus on other tasks.

“The result is increased productivity,” she said. “We are also using RPA to think about financial management system modernization efforts, as we are doing data cleansing and data conversion. We are using RPA to do a lot of data crunching to help us to sort through data and prepare it for migration.”

But bots can make mistakes, and they also have limited functionality.

“Right now our robot just produces a report for the end user, and the user still has to reconcile data,” Gaublomme said. “We want to use machine learning (ML) so the system can adapt and learn to make better decisions. We also want to use ML for the detection of fraud. There are a couple federal use cases on ML. The IRS is using it to quickly and accurately detect fraudulent tax returns.”

The goal, she added, is a 24/7 digital workforce and improved user experience.