ANN-based Command and Control Seat Allocation Optimization with Simulation Data

In previous studies of the command and control (C2) modeling and optimization, the C2 process tends to be simplified to a large extent, resulting in a relatively large deviation to actual operation. The paper adopts a simulation method and surrogate technology to transform the black box C2 system into a white box model and try to optimize the C2 seat allocation problem by maximizing the C2 efficiency. Firstly, the dynamic C2 process is modeled based on the simulation tool of ExtendSim, by mining key elements and relations in C2 process. Secondly, with training data generated from the ExtendSim, the artificial neural network (ANN) based surrogate model is adopted to approximate the relation between allocation solution inputs and efficiency outputs. Then, by treating the trained ANN as the objective function, the non-dominated sorting differential evolution (NSDE) algorithm is used to obtain the Pareto set of the problem. Finally, a case is studied to verify the feasibility and effectiveness of the proposed idea and methods.