This research addresses the challenge of allocating limited resources across offensive cyber operations (OCO) and defensive cyber operations (DCO) in the U.S. Navy (USN), where U.S. Fleet Cyber Command/Commander, U.S. 10th Fleet (FCC/C10F) receives 4% of the total USN funding. With cyber capabilities playing a larger role in modern warfare, it is vital to understand how to optimize OCO, defensive cyber operations–responsive action (DCO-RA), and defensive cyber operations–internal defensive measures (DCO-IDM). This study applies influence diagrams and neural networks (NNs) to model the effectiveness of Blue Force cyber operations under varying funding levels, intelligence quality, and adversarial strength. A Deep Q-Network (DQN) was developed to simulate a range of operational scenarios and adversary behaviors, enabling strategic decision making under uncertainty. Results show how different funding strategies affect operational outcomes and identify allocations that maximize mission effectiveness. The model helps decision makers visualize tradeoffs and anticipate adversary effects, offering a tool for refining cyber investment strategies. Future work should integrate a more dynamic adversarial response and real-time intelligence updates to further enhance decision making in an uncertain environment.

Distribution Statement A. Approved for public release: Distribution is unlimited.

This project was funded in part by the NPS Naval Research Program.

Lieutenant, United States Navy

NPS Naval Research Program