Project

Using Natural Language Processing via Machine Learning techniques, this research project aims to provide a data-driven computational model and working prototype to support the planning cycle by injecting relevant threat-based intelligence and operational scenarios into the more traditional capabilities-based planning. 

Capabilities-based planning attempts to break down traditional enterprise-level stovepipes during system development to stem the effects of a varied adversary. However, this approach tends to isolate the technical development community from strategic and tactical operational considerations. Injecting relevant threat-based intelligence and operational scenarios into the capability-based planning approach will better inform the technical communities charged with developing future weapons systems. This approach has been piloted in late 2016 at the U.S. Army Combat Capabilities Development Command Armaments Center (CCDC AC) in the armament-systems domain. The researchers shall focus on creating a computational framework and a related system prototype that will:

1. Replicate the aforementioned process developed at CCDC AC in 2016 to validate this notional computational architecture.
2. Enhance the visualization and analytic capability to allow rapid, high fidelity decision making.
3. Introduce additional parameters and variables to further refine the decision making framework.

Real-world scenarios will be modeled to project evolving threats, doctrine, partner force interoperability, and other operational environmental conditions (political, military, socio-economic, information, infrastructure, physical environment).

The research project is distributed in 2 years, where Year 1 focused on acquiring the logic used in the planning process, creating a corpus to be used for the data/text mining, developing the required components, and creating the proof of concepts for the systems. The proof of concepts was delivered to the Sponsor at the end of Year 1 (June 2019). 

During Year 2 (June 2019 – June 2020), the team will focus on adding functionalities to the proof of concepts and evolving them into a working prototype. Majority of the activities in Year 1 will be expanded and enhanced in Year 2 to make the systems more effective for the Sponsor.

When fully realized, this framework will contain a catalog of threats, operational scenarios, and tactics, techniques, and procedures (TTP) to be used in future analysis. The framework shall be interactive, scalable, and accessible by a wide range of users. Further, the framework will be capable of performing trades at multiple levels of fidelity depending on needs (lower fidelity for initial rapid assessment of wide-search spaces and higher fidelity for more focused searches). The framework should be configured in such a way to allow real-time updates based on feedback from the field and/or other sources.

• SERC-2019-TR-010-Meshing Capability and Threat-based Science and Technology (S&T) Resource Allocation

• Poster - RT 166: Formal Methods in Resilient Systems Design using a Flexible Contract Approach
• Poster - RT 182: Enterprise Systems-of-Systems Model for Digital Thread Enabled Acquisition
• Poster - Formal Methods in Resilient Systems Design using a Flexible Contract Approach
• Poster - Meshing Capability and Threat-based Science & Technology Resource Allocation