Simulating a logistics enterprise using an asymmetrical wargame simulation with soar reinforcement learning and coevolutionary algorithms

We demonstrate an innovative framework (CoEvSoarRL) that leverages machine learning algorithms to optimize and simulate a resilient and agile logistics enterprise to improve the readiness and sustainment, as well as reduce the operational risk. The CoEvSoarRL is an asymmetrical wargame simulation that leverages reinforcement learning and coevolutionary algorithms to improve the functions of a total logistics enterprise value chain. We address two of the key challenges: (1) the need to apply holistic prediction, optimization, and wargame simulation to improve the total logistics enterprise readiness; (2) the uncertainty and lack of data which require large-scale systematic what-if scenarios and analysis of alternatives to simulate potential new and unknown situations. Our CoEvSoarRL learns a model of a logistic enterprise environment from historical data with Soar reinforcement learning. Then the Soar model is used to evaluate new decisions and operating conditions. We simulate the logistics enterprise vulnerability (risk) and evolve new and more difficult operating conditions (tests); meanwhile we also coevolve better logistics enterprise decision (solutions) to counter the tests. We present proof-of-concept results from a US Marine Corps maintenance and supply chain data set.