Optimization Of Low-Carbon Multimodal Transport Routes For Containers Considering Uncertain Demand

Container multimodal transport integrates many advantages and is one of the main popular forms of transport,which helps to realize the development strategy of China’s strong transportation and enhance the core competitive advantage of our economy,and is also an important means to reduce the operating costs of road transport.Due to the influence of various factors inside and outside the market,the demand for multimodal transport will change dynamically,which will have an impact on the total transport cost and the choice of transport paths,while low-carbon transport occupies a very important position in the low-carbon economy and will be an important trend in the development of China’s transportation industry in the future.Therefore,under the premise of uncertain demand,the research on low-carbon multimodal transport path optimization is of great practical and theoretical significance for economic efficiency and low-carbon development.In this paper,the content of container low-carbon intermodal transport path optimization based on demand uncertainty conditions is divided into the following three main aspects:(1)Multimodal transport model construction and transformation.Firstly,we take the total cost of transportation,transit cost,transportation carbon emission cost,transit carbon emission cost and the sum of penalty cost generated at nodes as the total cost in the transportation process,and establish a deterministic model of multimodal transportation with the total cost minimization as the objective function and the path transportation capacity,total transportation time limit,node flow balance,and transportation continuity as the constraints.In the case of demand uncertainty,it is analyzed that demand uncertainty is involved in both the objective function and the constraints,and the Bertsimas robust optimization method is used to characterize the demand uncertainty parameters and linearize the model in the uncertainty model transformation,while the transformation of the demand uncertainty model is realized with the help of the pairwise theory.(2)Design genetic particle swarm algorithm and improve it.Based on the advantages of fast convergence and simple understanding of particle swarm algorithm,we choose particle swarm algorithm as the basic solution algorithm,and make optimization for inertia weights and learning factors in particle swarm algorithm.In order to prevent premature convergence,the operations of selection,crossover,and variation in genetic algorithms are introduced,and the elite selection strategy,arithmetic crossover,and variation methods based on particle swarm ideas are chosen respectively,and the genetic particle swarm algorithm solution model is finally designed.(3)Results analysis based on realistic examples.The multimodal transportation network from Shenyang to Changsha,which contains 20 cities,is used as the object of study.The performance of the three algorithms is first verified through the examples,and then the sensitivity analysis of the deterministic model is carried out for the arrival time limit.Secondly,for the uncertainty model,sensitivity analysis is performed by taking the number of uncertain budget,the size of disturbance fluctuation,and the value of carbon tax.Through the result analysis,it is verified that the uncertainty model has good robustness and also provides suggestions for multimodal transport enterprises.Multimodal transportation integrates the advantages of various modes of transportation,and the reasonable planning of cargo transportation paths and transportation solutions is of great significance to achieve lower transportation costs,improve efficiency and perfect the transportation system.