The Application Of Improved Genetic Algorithm On Cold Chain Logistics Vehicle Routing Problem

In recent years,with the promotion of consumption upgrading and the development of e-commerce,the demand for cold chain products has been increasing.However,in China,the distribution of cold storage is uneven,cold chain insulation technology still has gaps with international standards,the cold chain cost is too high,and the cold chain talent resources are short.These problems has brought huge obstacles to the development of cold chain logistics.Cold Chain Logistics is a systematic project of ensuring the product quality and as well as reducing the product loss,which the refrigerating products are always in the low-temperature environment from the production,storage,transportation and sales to the consumers.The distribution of cold chain logistics mainly includes agricultural products,fresh aquatic products,processed foods and special commodities,which are characterized by perishability.Therefore,cold chain logistics has high requirements for timeliness.In the cold chain logistics,rational planning of the driving route for the refrigerated truck can not only shorten the delivery time,reduce the cost of the enterprise,but also improve the customer satisfaction.Both in theory and in reality,there is great research significance.The main work is done as follows:(1)Discussed the cold chain logistics vehicle routing optimization problem,based on the consideration of fixed cost,transportation cost,cargo damage cost,energy cost and penalty cost,and then established cold chain logistics vehicle routing optimization model.(2)The basic mutation operator selects one or more genes randomly from the coding string of chromosomes in the population,which makes the algorithm poor population diversity and more converge slowly.Thus,two improved genetic algorithms was proposed.On the basis of analyzing seeker optimization algorithm,used the uncertainty reasoning behavior and the nearest neighbor strategy to improve the mutation operator in the genetic algorithm,proposed Seeker Genetic Algorithm(SOA-GA).Fireflies individuals can be introduced into the mutation operator of genetic algorithm,mutual attraction between fireflies can reduce the randomness of the selection of interchangeable genes,that is firefly variation.At the same time,in order to increase the diversity of the population and prevent it falling into a local optimum,variable neighborhood disturbance mechanism is introduced in the process of firefly variation,proposed firefly Genetic Algorithm(FGA).Verified the effectiveness of the two improved algorithms by a typical test function and used to solve cold chain logistics vehicle routing optimization model,which reduces the distribution cost.(3)Used Beijing-Tianjin-Hebei Metropolitan about 13 cities as the simulationexample with the “next day”distribution model of JingDong,we got the lowest distribution plan of cold chain in Beijing-Tianjin-Hebei Metropolitan solved by three algorithms.Compared with the optimization results of simple genetic algorithm,the seeker genetic algorithm is the best and the firefly algorithm is suboptimal.Finally,according to the simulation results,we put forward corresponding suggestions.