Avum, Inc. is revolutionizing Department of Defense (DoD) procurement through key government collaboration. A recent research paper was co-authored by Samantha Nangia, a Senior Program Analyst and Procurement Technology Transformation Lead, and Thomas Wardwell, a Contracting Officer and Financial Manager, both from the office of the Assistant Secretary of the Navy for Research, Development and Acquisition (ASN (RDA) DASN (P)), along with Avum's CEO, Randall Mora, and data scientist Carlos Parada. The research explores the successful integration of Generative AI to enhance the quality and efficiency of DoD acquisition and contracting processes.
The paper, titled "Leveraging Generative AI for Validating the Quality of DoD Acquisition Packages and Contract Documents," details a groundbreaking study on the use of a secure AI model, NIPR GPT, to automate and improve the validation of complex procurement documents. This is a critical area for the DoD, which operates within a highly regulated environment governed by a complex web of regulations like the Federal Acquisition Regulation (FAR) and the Defense Federal Acquisition Regulation Supplement (DFARS).
Tackling a Complex Challenge
The DoD's contracting process is notoriously intricate, requiring meticulous validation to ensure compliance and accuracy. Contracting officers face the significant challenge of navigating these complex and frequently updated regulations, a process that is often labor-intensive and prone to human error. This can lead to delays, increased costs, and compliance risks.
An Innovative AI Solution
The research demonstrates the immense potential of Generative AI, and specifically the use of Retrieval-Augmented Generation (RAG), to address these challenges. By fine-tuning the AI on DoD-specific datasets, the researchers were able to create a powerful tool that can:
- Automate Compliance Checks: The AI can rapidly check documents for compliance with FAR, DFARS, and other regulations, identifying missing clauses and other potential issues.
- Enhance Accuracy: The system significantly reduces the risk of human error, leading to more accurate and reliable contract packages.
- Increase Efficiency: By automating many of the manual review tasks, the AI-powered system can dramatically accelerate the entire procurement process.
Proven Results and a Promising Future
The study's findings are compelling. In a case study involving a major aircraft procurement, the NIPR GPT model successfully identified missing compliance language and ambiguous descriptions, streamlining the approval process. The research also showed a high degree of alignment between the AI's findings and those of human reviewers, with the AI's performance improving with each iteration.
"This research highlights the transformative role of generative AI in supporting DoD contracting officers," the paper states, "by providing reliable, secure, and efficient tools for package validation".
Avum is proud to be a part of this innovative work, which showcases our commitment to developing cutting-edge solutions for our clients. The successful integration of AI into DoD workflows illustrates a significant step forward in modernizing procurement practices, improving accuracy, and ensuring compliance in a highly regulated environment. The research is ongoing, with future work focused on further refining the AI models and expanding their capabilities to continue to enhance mission readiness for the DoD.
Explore the Research
For more details, you can access the original research paper and the accompanying presentation materials on the Naval Postgraduate School’s Acquisition Research Program website: Leveraging Generative AI for Validating the Quality of DoD Acquisition Packages and Contract Documents