Material Dispatch For Public Emergencies Based On Improved Hybrid Frog Leaping Algorith

The shuffled frog leaping algorithm is a meta-heuristic algorithm with the advantages of few parameters,simple structure and strong global search capability.Public emergencies are natural disasters,public health events and social security events that cause serious social hazards and require emergency measures to be taken.According to the characteristics of sudden,continuous and uncertain of public emergencies,this paper designs single-objective shuffled frog leaping algorithm,dynamic multi-objective shuffled frog leaping algorithm and dynamic interval multi-objective shuffled frog leaping algorithm,and applies them to the material scheduling problems under three kinds of public emergencies respectively,and the research contents are as follows.Firstly,based on the differences in the damage situation in the disaster areas and the limitation of the number of vehicles,a mathematical model of emergency material scheduling that considers the degree of rescue urgency and the splittable supply of demand is established.To solve the model,a single-objective shuffled frog leaping algorithm with multiple information resources learning is designed.The algorithm generates new individuals through a multi-individual greedy crossover strategy to expand the search space of the algorithm;it uses local information crossover to implement deep mining on the worst individuals in the subgroup to improve the ability of the algorithm to jump out of the local optimum.The performance of the improved strategies and the proposed algorithm are verified on Wenchuan earthquake emergency material scheduling instance and 10 synthetic instances.Compared with five state-of-the-art algorithms,the proposed algorithm is able to search for vehicle routing solutions with shorter vehicle travel distances and less delayed supply in the disaster area for problems of different scales.Secondly,considering the differences in testing unit price and testing capacity of testing organizations,a multi-trip and multi-depot vehicle routing problem with dynamic periodicity characteristics is established for the transfer of nucleic acid specimens using cross-depot collaborative transportation.A knowledge-guided dynamic multi-objective shuffled frog leaping algorithm is proposed.The response mechanism of historical information reuse is used to provide a good search starting point for the algorithm;the greedy crossover and genetic recombination-based leaping rules are designed by combining heuristic information to realize the coarse-grained search of the algorithm;the objective-driven enhanced search mechanism is used to realize the fine-grained search of the algorithm.The comparison results with five state-of-the-art algorithms on the nucleic acid specimen transfer instance in Nanjing Jiangbei New District show that the proposed algorithm can provide a set of non-dominated transfer solutions with lower cost and shorter testing time in each scheduling period.Finally,Considering the variation of vehicle travel speed,two types of dynamic events are introduced to construct a medical waste collection under the COVID-19 pandemic with dynamic uncertainty characteristics.To solve the model,a two-stage scheduling mechanism is designed.In the static optimization stage,the interval multi-objective shuffled frog leaping algorithm is used to solve the waste collection routing at the initial moment through the knowledge-guided interval multi-objective shuffled frog leaping algorithm.The algorithm introduces the possibility degree of interval number to achieve individual decoding of travel speed as an interval number and non-dominated sorting of the population.In the dynamic optimization stage,the waste collection routing is adjusted by the neighborhood search based on problem features in order to improve the real-time processing capability of dynamic requests.The improved strategies and the performance of the proposed algorithm were verified on Nanjing medical waste collection instance and eight synthetic instances.In comparison with six state-of-the-art algorithms,the proposed algorithm can search for a set of medical waste collection solutions with lower scheduling costs and less risk.