The Homeland Security Department is looking for automation, machine learning and artificial intelligence tools to get its financial systems where they need to be.
As part of the ongoing Financial Systems Modernization Program, the department has been working to move away from paper-based invoicing. But to fully realize that goal, the program will need high-quality tools capable of scanning and digitizing paper documents, according to a sources sought notice posted to beta.SAM.gov.
The Office of Management and Budget issued a mandate in 2015 for all agencies to move to electronic invoicing within three years. While DHS financial modernization moved away from paper invoicing, on the whole, the efforts did not meet the minimum requirements, according to the notice.
The document calls out two tools purchased as part of the modernization program—Kofax Capture and MarkView—neither of which “include automated technologies, such as machine learning and artificial intelligence.”
“Based on the volume of transactions processed by FINCEN,” the Coast Guard’s central financial processing center, “and without these additional automated capabilities, compliance with federal mandates and requirements of the Prompt Payment Act and GAO’s Standards for Internal Control in the Federal Government—known as the Green Book—may be impacted during [Coast Guard’s] move” to the new system, the notice states. “Utilizing current technologies such as machine learning, artificial intelligence, as well as other technologies will enable processing volumes transactions in a more efficient manner.”
The sources sought notice is looking for information on vendors able to support the engineering of financial processing systems that can function across as many DHS components as possible.
Along with the Coast Guard, the joint program management office is also currently working with the Countering Weapons of Mass Destruction office and the Transportation Security Agency—“collectively known as the ‘Trio,’” according to the notice. The program expects to add two more implementations this fiscal year and next, and potential vendors should be ready to add those to their workloads.
The contract will cover several support tasks, including:
- Data capture and optical character recognition for turning paper invoices into machine-readable documents.
- Automated workflow and robotic process automation.
- Process analytics and dashboards.
- Creating and managing test and development environments.
- Training for initial user groups.
Each task carries a list of specific deliverables detailed in the notice.
The contract is expected to run for a base period of one year, with an additional one-year add-on option.
Responses to the sources sought notice are due by noon on Feb. 8.