Application And Research Of Improved Hybrid Ant Colony Algorithm In Low Carbon Cold-chain Logistics

In recent years,the effect of the COVID-19 has inspired the development of fresh food e-commerce,cold chain distribution and other related industries to some extent.With the increasing problems such as global warming,however,low-carbon issues are receiving more and more attention from countries and governments.The industry of cold chain logistics and distribution is a major consumer of energy and a major channel of carbon dioxide emission,so it has become an industry trend to reduce the energy loss and carbon emission in distribution.In the actual logistics and transportation process,cold chain delivery vehicles may suffer from traffic jams and other disruptions during the delivery operation,resulting in more energy consumption and more carbon dioxide emissions than expected.Therefore,it is especially important to optimize the cold chain logistics distribution disruption management under the carbon emission mechanism,reduce the carbon emission of vehicles,and find green development and win-win benefits for companies.The main researches of this paper are as follows.First of all,a Deep Learning-based traffic congestion prediction model is built to predict the traffic flow in a short period of time for congestion periods based on carbon emission strength mechanism using Long Short-Term Memory Network(LSTM)in Deep Learning.After that,the initial scheduling model of low carbon cold chain logistics distribution is established.Taking the traffic congestion event as the disturbance event in cold chain logistics distribution and considering three objectives of carbon emission cost,path deviation and comprehensive cost deviation,constructing a low carbon cold chain logistics distribution disruption management model under the carbon emission strength mechanism.Based on theoretical research related to disruption management,a disruption metric combining congestion prediction model and carbon emission strength mechanism is proposed.Secondly,an Improved Hybrid Ant Colony Algorithm(IHACO)is proposed,which combines the Ant Colony Algorithm and Simulated Annealing Algorithm to avoid the phenomenon of stagnation or premature maturity of the Ant Colony Algorithm,to make the algorithm break out of the partial optimum,to change the state transfer rules and update the pheromone update way,introducing Chaotic Mapping to improve the efficiency of the algorithm,to quicken the algorithm’s convergence speed and to improve the pheromone volatility speed,so as to improve the performance of the algorithm and to solve the problem of low carbon logistics distribution disruption management.At last,the Improved Hybrid Ant Colony Algorithm(IHACO)and the low-carbon logistics distribution disruption management model are validated by combining the relative data from the references.The convergence speed of the Improved Hybrid Ant Colony Algorithm and the feasibility of the improved strategy are verified by comparing the classical arithmetic experiments and the execution efficiency of different algorithms.