code

In the process of solving the 2E-VRP problem, the large-scale destruction and repair algorithm is used to ensure that the algorithm does not fall into the local optimal solution. The process of the initialization process uses the greedy strategy to cluster the customers. The large-scale destruction process is to randomly remove the customer nodes on the satellite into the customer pool. The repair process is based on the reciprocal of the customer's distance to each customer in the customer pool. Gambling Select the satellite to which the customer belongs and engage in greedy insertion. For the second layer of path planning, you need to use multiple search operators, such as random sequence reversal exchange operator, crossover operator, damage and repair operator, and crossover operator variants, etc. to improve the artificial bee group algorithm Of the local search ability. To ensure that the global optimal situation can be found, the neighborhood of large-scale search. The improved artificial bee colony algorithm incorporates the idea of ​​simulated annealing and improves the global optimization ability of artificial bee colony algorithm. For the artificial bee colony algorithm, the combination of global optimization ability and local optimization ability improves the possibility that the algorithm can find a better solution than the existing method. Multi - operator artificial bee colony algorithm, which extends the search range of the food source 's neighborhood, and more possibilities to find the global optimal solution. The experimental results show that the algorithm can get better path planning results