Research On Computational Performance Of Harmony Search Algorithm For Solving Multi-objective VRPTW

Vehicle Routing Problem(VRP)is one of the most challenging problems in the field of combinatorial optimization problem,and it is a common problem in the application of practical path optimization such as logistics distribution.In this paper,the Multi-objective vehicle path problem(Multi-objective Vehicle Routing Problem with time Window)is studied,aiming at minimizing the number of vehicles and the distance of the vehicle,and a Pareto optimal solution is solved for decision makers to make their own decision support based on the solution set.The Intelligent optimization algorithm itself has strong global search ability,and many related algorithms are used to solve the problem of vehicle path and its variants.In this paper,a hybrid intelligent algorithm for solving multi-objective VRPTW is studied,which mainly does the following work:(1)For solving the combinatorial optimization problem of this type,the first choice of this paper is to use the improved multi-objective harmony search algorithm(MOIHSA),HS algorithm has a strong multi-objective optimization characteristic,which is very suitable for solving multi-objective VRPTW.In the algorithm,the structure of a good and sound memory is designed to preserve the evolving harmonies in a set of optimization processes,and the collation of harmonies in the acoustic memory is set up to facilitate the update operation of the harmony memory library.Secondly,the new harmony generation process in the original HS algorithm is improved,and the pitch is selected from the harmony memory library or the harmony search range by introducing the variable neighborhood search,and the tone of the selected self-tuning memory is fine-tuned through dynamic and acoustic bandwidth.Finally,for the key parameters such as the retention probability HMCR of the harmony memory and the fine-tuning of the disturbance probability PAR,the dynamic parameters are used to optimize the depth and breadth of the dynamic jamming algorithm,according to the optimization characteristics of the algorithm execution process.Through the experimental simulation results,it can be seen that it is feasible and effective to use the improved HS algorithm to solve the multi-objective VRPTW.(2)According to the above experimental results,it can be seen that MOIHSA has some shortcomings,such as relying heavily on the initial solution,which shows that the algorithm has multiple differences in search results,and the previous optimization ability of the algorithm is not strong,and the new solution component is randomly generated.Too much sex.In response to these shortcomings,this paper proposes an improved multi-objective HS algorithm(GA-MOIHSA)mixed with genetic algorithm(GA)to solve multitarget VRPTW.In the hybrid intelligent algorithm,the candidate set of HM neutralization sound is generated by first utilizing the high concurrency characteristics of GA and the advantages of the algorithm to improve the solution in the initial stage.That is,through GA continuous selection,crossover and mutation operations,after a certain number of iterations,the best set of solutions is selected from the population as the initial solution of the MOIHSA in the HM.GA also has its own shortcomings,such as high diversity of population requirements and slower convergence in the later stage.By mixing with IHSA,we can learn from each other.According to the experimental simulation results and the comparison experimental results,it can be seen that GA-MOIHSA can make up for the shortcomings of MOIHSA and improve the performance of MOIHSA.