| With the rapid development of social economy,the mileage of road,railway and other transportation is also increasing rapidly,and various modes of transportation are gradually developing towards integration and integration.Public rail combined transport is one of the main modes of multimodal transport.With the increasing volume of traffic,the existing problems are becoming more and more prominent.Unreasonable public rail combined transport scheme will not only lead to the increase of transport costs,but also lead to the loss of highvalue customers.Therefore,this paper combined with the relevant theories of customer classification and route optimization to study the railway combined transport,ensure the punctuality rate of high-value customers and reasonably optimize the railway combined transport route.In this paper,based on the RFM model of consumption proximity,consumption frequency and consumption quota in customer relationship management,combined with the customer characteristics of railway combined transport,eliminate the influence of multicollinearity,and put forward a four-dimensional customer classification index system,which combines the length of customer relationship with the consumption proximity,consumption frequency and average consumption quota representing customer value.Through expert scoring method and analytic hierarchy process to determine the weight coefficients representing each dimension of customer value,the public railway combined customer value evaluation model is built.From the perspective of customer relationship length and customer value,the K-means algorithm is used to cluster customers,and the data of the survey is analyzed by practical examples.On the basis of considering the classification of customers,the railway transportation cost,road transportation cost,transfer cost and time penalty cost were analyzed in the railway combined transportation,aiming at the lowest total cost,the route optimization model of railway combined transportation was established.A hybrid adaptive genetic simulated annealing algorithm was designed,and a genetic algorithm with good universality and robustness was selected to increase the adaptive crossover probability and adaptive mutation probability on the basis of which the convergence rate of the population was accelerated at the early stage of evolution and the diversity of the individual population was increased at the late stage of evolution.A new solution acceptance mechanism of simulated annealing algorithm is added to accept inferior solutions to a certain extent to avoid the algorithm falling into local optimal prematurely.The effectiveness of the hybrid adaptive genetic simulated annealing algorithm is verified by comparing the average solving quality,stability and average iteration times of the three algorithms through several groups of examples with different scales.Finally,taking the traffic network around the intersection of Jiaozuo-Liuzhou Railway and Longhai Line as the research object,based on the customer classification results,the hybrid adaptive genetic simulated annealing algorithm is used to solve the scientific and reasonable distribution scheme of railway combined transport,which verifies the effectiveness of the model and provides a scheme for the transport resource allocation decision of railway combined transport enterprises.A controlled experiment was designed and solved several times without considering the customer classification.It can be concluded that: considering the classification of customers,the combined route optimization scheme of public railway transportation solved by the hybrid algorithm in this paper reduces the average total distribution cost by 2.2%compared with the adaptive genetic algorithm;it ensures the punctuality rate and satisfaction of high-value customers,reflects the strategy of customer classification priority,and provides a reference for the transportation resource allocation decision of public railway combined transport enterprises. |
PDF Full Text Request