Application Research On Emergency Grain Transportation Problem Based On Differential Ant Colony Algorithm

In recent years,natural disasters have been frequent both domestically and internationally,with various disasters such as volcanic eruptions and earthquakes directly affecting people’s lives and health.At the same time,China is also a country prone to natural disasters.Therefore,in the event of emergencies,how to quickly respond to the needs of disaster affected locations and transport emergency supplies to various demand points through vehicles is an important issue that society and relevant researchers need to solve.Grain is currently one of the most important disaster relief supplies.It is an emergency grain transportation problem to transport grain to various demand points through vehicle transportation while minimizing the total transportation cost while meeting the time window of the affected point.Through research on the emergency grain transportation problem,it was found that the basic model of the emergency grain transportation problem is the time window vehicle routing problem model.At present,a large number of studies both domestically and internationally have shown that ant colony algorithm has good performance in solving time window vehicle routing problems.However,when solving the time window vehicle routing problem,the ant colony algorithm is easy to enter the local optimal situation in the later stage of the algorithm.At the same time,the path selection strategy and the path choice function both have shortcomings.In order to solve the shortcomings of the ant colony algorithm,the improved selection strategy is deterministic and random sexual selection in two different ways.The improved path choice function takes into account the width of the time window and the width of the waiting time,which can prevent the algorithm from prematurity and increase the importance of customers.At the same time,in order to avoid the problem of traditional ant colony algorithms falling into local optima in the later stage,this paper mixes differential evolution algorithm with traditional ant colony algorithm,uses mutation and crossover operations to increase the search range,and ultimately improves the quality of the solution.The experimental part was solved by Solomon examples.The specific methods are as follows:1.Two basic algorithms were used to solve the c101 test set,and the performance characteristics of the ant colony algorithm and differential evolution algorithm were obtained,proving the feasibility of combining the two algorithms.2.Through three sets of comparative experiments,experimental results from different datasets were obtained.The comparative summary and analysis of the experimental results showed that the differential ant colony algorithm had good solving ability and algorithm feasibility.At the same time,the differential ant colony algorithm was compared with traditional ant colony algorithms and genetic ant colony algorithms in literature,and the optimal solution of the total cost respectively was 9045.08,10915.22,10100.36.This proves that the differential ant colony algorithm improves the quality of understanding compared to the other two algorithms,can obtain better path delivery solutions,and is also an effective solution.