Training & Events

2021 TTS Innovation Challenge Winner!

2021-Innovation-Challenge-Winner-Badge.png
Financial Obligation Root Cause Explorer (FORCE)
NSF developed the Financial Obligation Root Cause Explorer (FORCE), which uses a fully operational artificial intelligence and machine learning technology stack of data aggregation, visualization, natural language processing/text analytics, and dashboard platforms to quickly analyze NSF's portfolio of 50,000 open grants. This analysis identifies awards at risk of atypical spending, cluster analyzes unstructured text within proposal data and progress reports, and uncovers common themes in the underlying research activities. FORCE allows users to synthesize qualitative data in near real time across the entire portfolio and provide business intelligence for risk management. Previously, this type of work would require the time-consuming process of researching individual awards within the portfolio.

 

Prior Winners

2020-Innovation-Challenge-Winner-Badge-(2).png

National Science Foundation Team: Statistical Predictive Model

NSF developed a statistical predictive model and web scraper, using data from the Federal Audit Clearinghouse, agency systems, and open-source development platforms, to help forecast which NSF awardees may present a relatively higher risk of improper payments. This model minimizes manual burden through automation and provides a more quantitative view of risk, allowing NSF to better allocate its resources in risk assessment and grant monitoring activities. Using open-source code and free technologies, the bulk of this model is transferable to other agencies. It can be customized to accommodate specific program risk factors and help address Payment Integrity Information Act of 2019 compliance requirements.

Bureau of the Fiscal Service, Digital Workforce Support Team: Robotic Process Automation

The team successfully piloted the use of RPA on a variety of routine, rules-based processes and decided to move forward with full-scale implementation, given the potential of RPA to significantly reduce manual processing as well as increase efficiency and accuracy. The bots interact with financial management, procurement, travel and HR systems, and they run unattended, freeing up staff to shift to other valuable work. Additional results include a reduction in processing time by up to 99%, increased compliance (100% accuracy) and improved staff morale (reduction of repetitive work and over time.)