Research On The Multi-Objective Cold Chain Logistics Vehicle Routing Problem Based On The Genetic Algorithm-Ant Colony Algorithm

With the increasing of productivity and the rising of fresh new retail,the consumption of fresh agricultural products is increasing,and people pay more attention to its quality and distribution service.Fresh agricultural products are easy to rot at room temperature and cause additional loss,it is necessary to adopt the cold chain logistics to transport and distribute.The cold chain logistics in our country have the deficiency of inadequate facilities,high distribution cost,serious loss,low punctuality and high rate of empty loading.It is necessary to improve status quo.So this paper study the multi-objective fresh agricultural products cold chain logistics vehicle routing problem with capacity and time window constraints,and build the lowest total cost and highest customer satisfaction as the optimization goals of the multi-objective optimization model.The differences between the transport process and unloading process,and the effect of goods change in the delivery process are considered in the definition of refrigeration cost and damage cost.Customer satisfaction reflects customers' evaluation to the pick up or delivery time.In order to solve the problem of multi-objective optimization,the method of epsilon constraint is adopted.Combining the advantages of genetic algorithm and ant colony algorithm,design genetic algorithm-ant colony algorithm,introduce the crossover operator and mutation operator of genetic algorithm into ant colony algorithm.Firstly,generate the initial population by the ant colony algorithm,then impose crossover operation and mutation operation,and optimize the state transition rule and total pheromone,thus reduce the prematurity of the algorithm and improve the convergence speed and solution quality of the algorithm.To verify the validity of the model and algorithm,this paper solve the actual example,and compare the results obtained by this algorithm with the results obtained by genetic algorithm and ant colony algorithm.By comparing,we can find the optimal results obtained by this algorithm is more excellent than the results of genetic algorithm and ant colony algorithm,and can find the optimal solution in the least number of iterations.Showing that the model is reasonable,the algorithm designed in this paper can improve the convergence speed and solution accuracy.In addition,the experimental results show that for the instance of this paper,the total cost is the highest when customer satisfaction is 100%.When the customer satisfaction is reduced to 95%,the total cost is reduced and the decrease is larger.When customer satisfaction is reduced from 95% to 80%,the total cost is slightly reduced,but the decrease is small.When customer satisfaction is less than 80%,the total cost is stable and no longer decreases.It shows that for enterprises with customer satisfaction as the most important measure,and costs as the second factor,ensuring customer satisfaction 100% is more consistent with the business strategy of the enterprise.For enterprises with equal customer satisfaction and costs,ensuring customer satisfaction 95% is more consistent with the business strategy of the enterprise.For enterprises with costs as the most important measure,and customer satisfaction as the second factor,ensuring customer satisfaction 80% is more consistent with the business strategy of the enterprise.