Research on cross regional emergency material scheduling algorithm based on seed optimization algorithm

• Published in: Scientific reports Vol. 13; no. 1; pp. 21968 - 12
• Main Authors: Li, Jinbao, Cui, Lichong, Chu, Huayu, Su, Lei, Wang, Junsheng
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 11.12.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166; 639/705; Algorithms; Costs; Data processing; Decision making; Deep learning; Disaster relief; Disasters; Earthquakes; Efficiency; Emergency preparedness; Humanities and Social Sciences; Hurricanes; Integer programming; Linear programming; Logistics; Methods; multidisciplinary; Objective function; Optimization algorithms; Planning; Regional planning; Scheduling; Science; Science (multidisciplinary); Tsunamis
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-023-47538-2

In order to improve the response capability of cross regional emergency material scheduling (CREMS), a CREMS algorithm based on seed optimization algorithm is proposed. Construct a segmented regional grid distribution model structure for CREMS, use a grid matching algorithm based on block link distribution to construct the optimization objective function during the emergency material scheduling process, use variable neighborhood search technology to solve the diversity problem of cluster optimization in CREMS, and combine seed optimization algorithms for combination control and recursive analysis in the emergency material scheduling process. Based on the combination of deep learning and reinforcement learning, the optimal route and configuration scheme design for CREMS process is achieved. The simulation results show that this method has better active configuration capability, better path optimization capability and stronger spatial regional planning capability for CREMS.