A study of aeromedical emergency scheduling based on a collaborative difference algorithm improved by reinforcement learning

Lei Song, Sisi Xiao, Longquan Liao, Wenlong Yan, Linjiang Zheng

Abstract: With the increasing frequency of random occurrence of large natural disasters, aeromedical emergency rescue has gradually become a major hot spot in today's society. Timely and efficient rescue can save more lives in the golden moment, so more and more scholars have launched an in-depth study on the scheduling of aviation emergency resources. The purpose of this paper is to reasonably plan out the aircraft scheduling programme under large-scale disasters and make full use of the rescue materials. Therefore, this paper establishes a hierarchical collaborative scheduling model for aeromedical emergency rescue, with the objectives of minimising the total aeromedical rescue time cost, maximising the total utility of the rescue mission and maximising the satisfaction of the rescue mission. In this paper, an improved differential algorithm (RL-DE algorithm) based on the reinforcement learning framework is proposed to solve the model, and simulated data are used to prove the model and algorithm proposed in this paper, and the experimental results show that the method proposed in this paper can generate the scheduling plan better.