Considering Carbon Emission For Multi-depot Heterogeneous VRP Model And Algorithms Research

According to the international energy agency,the transportation is the second largest contributor of carbon dioxide emissions,and almost three quarters of CO2 emissions are come from road transport.Therefore,it is necessary to reduce the carbon dioxide emissions in vehicle transportation.Nowadays,logistics and distribution network is increasingly developed,in order to expand business and timely to meet customer requirements,many companies have set up several depots,and provide a variety type vehicles to adapt different types of goods.In order to make the research more close to reality and have more practical significance,the model of considering carbon emission for multiple depot heterogeneous fleet vehicle routing problem is been studied.This paper has carried out the following research under the support of the National Natural Science Foundation of China(71571120,71271140,71471158),the Guangdong Provincial Natural Science Foundation(2016A030310074)and Guangdong Provincial Science and Technology Plan Project(No.2013B040403005):First of all,this paper constructed a new model—Considering carbon emission for multidepot heterogeneous VRP model.This paper studied different models of VRP.And for carbon emissions problem,this paper mainly considered the energy consumption and carbon trading mechanism.Several energy consumption calculation method were be analyzed,and comprehensive calculation method was been taken.For different types of vehicles,fixed costs will be different.The model also takes the constraints of the time window into account.A penalties would be generated if earlier or later than the time window.Therefore,the model constructed is mainly studied the multi-depot heterogeneous fleet VRP model with considering carbon emission factors,taking the carbon trading mechanism into consideration,and want to make the total cost be minimum.Secondly,the paper analyzed and proposed an improved algorithm to solve the model.Bacterial foraging optimization algorithm is a novel type of intelligent optimization algorithms,with parallel search,good at local search advantages.Single loop structure of the bacteria foraging optimization algorithm(SRBFO)reduces the computational complexity,with more good performance in convergence and optimization.Enhanced comprehensive learning particle swarm algorithm(ECLPSO)adopt a comprehensive learning mechanism that can greatly improve the accuracy of the solution.Based on the existing research achievements,this paper combined a SRBFO and ECLPSO,and constructed a new algorithm of single loop structure comprehensive learning bacterial foraging optimization algorithm(SRCLBFO).To validity SRCLBFO algorithm,several standard test functions of unimodal functions and multi-modal functions of were tested.Finally,combining with the instance,the algorithm were applied to the problem.The validity of the model and the improved algorithm are verified by the comparison of the experimental results.At the same time,combined with this instance,this paper also analyze the single depot,multi-depot,testified the advantage of multi depot;And compared to a single type fleet and heterogeneous fleet,demonstrated the effectiveness of heterogeneous fleet model.In addition,the effect of fluctuating of carbon trading mechanism of carbon price and carbon cap on the cost,carbon emission and distance were been analyzed.This paper enlarged the research of vehicle routing problem with considering carbon emission,and provide some reference to enterprise about low carbon transportation.