Air Force Operational Energy has developed a risk-based fuel logistics modeling and simulation tool. We need to pair this tool with AI/ML techniques to develop operational plans and practices as well as to improve strategic decisions about investment in fuel logistics networks.
Air Force operations depend on the availability of fuel, but the logistics network that supports our forces is vulnerable to disruption and is often out of our control. We have developed a risk-based modeling system that can optimize fuel network planning in the face of risk as well as simulate the impact of adverse effects on the fuel delivery in discrete scenarios. This model can be integrated into larger simulations that explore either the operation of the network during a contingency or the improvement of the network at strategic timescales.
We believe the risk-based model can be paired with AI/ML techniques to explore these simulations at scale. The techniques may reveal key insights into concepts of operation, trade-offs between different strategies, and opportunities to gain decision advantage. They may also form the basis of decision support systems that could inform future operational systems as well as long-term planning. We would like to test the feasibility of applying these techniques to the simulation in the context of a real-world decision application, with options from developing fuel distribution plans in Title X wargames to advocacy for strategic investment in logistics capabilities
