Research on Green Transport Mode of Chinese Bulk Cargo

Research ArticleResearch on Green Transport Mode of Chinese Bulk CargoBased on Fourth-Party Logistics

Jixiao Wu Yinghui Wang Wenlu Li and Haixia Wu

Department of Economics and Management Shijiazhuang Tiedao University Shijiazhuang050043 China

Correspondence should be addressed to Yinghui Wang wang112111163com

Received 4 September 2021 Revised 15 October 2021 Accepted 5 November 2021 Published 24 November 2021

Academic Editor Eduardo Lalla-Ruiz

Copyright copy 2021 Jixiao Wu et al is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Due to the problems such as the excessive proportion of road transport and extreme carbon emission situation of Chinarsquostransport structure adjustment this paper combines the fourth-party logistics with the bulk cargo green transport It is advancingthe adjustment of Chinarsquos bulk cargo transport structure using fourth-party logistics is paper improves the particle swarmoptimization algorithm to compare the integrated cost and carbon emissions of different bulk fourth-party transport networks toverify the benefits of the fourth-party logistics on bulk cargo transport networksrsquo cost reduction and emission reduction eresults show that using the fourth-party logistics model to promote the transfer of cargoes from road to rail can reduce theintegrated cost of the transport network reduce carbon emissions and achieve green transport

1 Introduction

Since the signing of the Paris Agreement China has beenactively engaged in global emission reductions and has takenthe initiative to assume responsibility for emission reduc-tions [1] According to the China Statistical Yearbook datathe transportation industry is one of Chinarsquos fastest-growingcarbon emissions industries Between 2000 and 2015 thetotal carbon emissions of Chinarsquos transportation industryincreased from 5822 million tons to 1901 million tons Itincreased by a full 226 times with an average annual growthrate of 79 [2] Reducing the carbon emissions in thetransportation industry is of considerable significance toChinarsquos modern integrated transportation system and greentransport with ldquohigh energy efficiency low pollution lowconsumption and low emissionsrdquo [2] Transport freightvolume is closely related to carbon emissions in 2019Chinarsquos road freight volume reached 416064 billion tonsrail freight volume has reached 438904 billion tons androad freight volume has reached 948 times the railwayfreight volume [2] According to the US Federal RailwayAdministration report the carbon emissions from roadtransport at the same distance are nearly four times that ofrail transport [3] e above data show that the high

proportion of road freight is an essential reason for the highcarbon emissions of Chinarsquos transportation industry If it cangive full play to the advantages of railways in reducingcarbon emissions and implement the ree-Year ActionPlan issued by the State Council of China to guide thetransfer of cargo from road transport to railway transport assoon as possible it will greatly benefit the realization ofChinarsquos sustainable development strategy and the devel-opment of green transport

To promote the adjustment of transport structure re-duce the carbon emissions of transport improve the inte-grated transport efficiency and reduce the cost of logisticsthe General Office of the State Council issued theree-YearAction Plan for Promoting the Adjustment of the TransportStructure (2018ndash2020) in October 2018 which proposed theaction of upgrading the capacity of railway transport andmade it clear that promoting the ldquoroad to railrdquo of bulk cargotransport is the main direction of transport supply-sidereform [4] In September 2019 Chinarsquos National Develop-ment and Reform Commission also cooperated with rele-vant departments and issued the ldquoGuiding Opinions onAccelerating the Construction of Railway Special Linesrdquowhich aim to solve the ldquofirst and last kilometrerdquo problem ofthe bulk cargo of railway transport and to provide solutions

HindawiJournal of Advanced TransportationVolume 2021 Article ID 6142226 16 pageshttpsdoiorg10115520216142226

to the distribution problem of ldquoroad to railrdquo [5] In additionthe State Railway Administration of China has taken anumber of measures in railway logistics infrastructurerailway transport prices and preferential transport policiesto deepen the reform of railway freight transport and ac-celerate the process of green transport

Recently the traditional bulk cargo transportation in-dustry has gotten rid of the low-price-first model Shippersrsquodemand for the safety convenience and delivery time oftransport is increasingly higher e transportation industryurgently needs a new model to improve the economicbenefit In order to optimize transport costs under low-carbon emissions the following innovations are made in thispaper (1) e fourth-party logistics promotes the rapidtransfer of bulk cargo from road transport to railwaytransport improves transport efficiency constructs an in-telligent transportation system and optimizes the integratedcost of bulk cargo transport (2) Unlike traditional bulkcargo transport this paper focuses on the transport processbetween multiple shippers and freight enterprises More-over the improved particle swarm optimization algorithmcompares the integrated cost and carbon emissions of roadand rail transport under different conditions within a limitedtime

e remainder of the paper is structured in the followingmanner Section 2 presents an overview of the relevantliterature focusing on green transport bulk cargo transportfourth-party logistics and algorithm design Section 3 buildsthe optimization model of bulk cargo green transport modebased on fourth-party logistics while Section 4 uses differentexperiments to illustrate the optimization effect of thefourth-party logistics of bulk cargoes on green transportSection 5 summarizes the findings and outlines potentialfuture research extensions

2 Literature Review

In order to promote the development of green transportresearchers have carried out much research on the low-carbon emission reduction of the transportation industryey analyzed the local transportation system through theCUTE matrix and pointed out that the implementation oflow-carbon policies and improvement of transportationtechnology are essential means to reduce the carbonemissions of the transportation industry [6] A model ofurban vehicle distribution with low carbon energy savingand low cost was constructed and genetic algorithms weredesigned to optimize vehicle distribution routes [7] Mea-sured the carbon emissions of the integrated logistics systemof the port and proved that the carbon emissions oftransport heavy equipment and logistics services accountfor a significant proportion of the transportation system [8]ey focused on the impact of carbon taxes on differenttransport modes and compared the carbon emissions andeconomic benefits of Chinarsquos railways roads and waterwaysthrough a general equilibrium model [9] e above re-searchersrsquo research on low-carbon emissions provides animportant idea for this article to explore green logisticsHowever there are still not many studies on promoting

green logistics in academia and it needs to be furtherimproved

In analyzing the characteristics of bulk cargo and pro-moting the green transport of bulk cargo researchers basedon the case of coal railway transport networks used bi-leveloptimization to explore the route of railway freight under thecondition of network interference [10] ey used a complexnetwork to evaluate the freight capacity of Chinarsquos road andrailway nodes and explored the differences in the trans-portation of bulk cargoes by road and railway [11] Combinethe bulk cargo price prediction model with neural networkalgorithms to improve railway freight volume predictionspeed and accuracy [12] ey also combined the TOPSISmethod to analyze the efficiency of bulk cargo transport inthe Brazilian railway system proposed some methods foroptimizing the rail transport of bulk cargoes and improvedrail freight efficiency [13] However researchersrsquo research onbulk cargoes mostly focuses on water transport and theresearch on rail transport of bulk cargoes needs furtherinnovation

In the research of transport informatization researcherscombined the views of 48 experts from 20 countries toexplore the supply chain management model of the futuretransportation industry and logistics industry from the fiveDelphi methods of energy emissions consumer behaviorfuture transportation mode future supply design and in-novation and concluded that a supply chain managementmodel that can help decision making information sharingand sustainability would emerge in the future and thefourth-party logistics could provide this management model[14] ey proposed a discrete flow model to describe thereverse logistics supply chain of remanufactured productswhich provided some ideas for the research of fourth-partylogistics [15] Constructed an ldquoN + 1 + Nrdquo fourth-partylogistics model based on Chinarsquos ldquoVAT Reformrdquo policy toachieve overall symmetry in logistics business informationand capital flows and explored the fourth-party logisticsprofit model [16] Combined with Internet technology itintegrated financial institutions consulting companies andother industries into the logistics industry chain to promotethe development of industry informatization and improvethe efficiency of logistics industry operation [17] Re-searchers integrated the government procurement infor-mation network through cloud technology to improvegovernment procurement efficiency and optimise supplychain management [18] e above researchersrsquo research onthe combination of logistics and big data provides manyideas for the research of fourth-party logistics in this paperHowever the existing logistics informatization researchmostly focuses on emergency logistics and the research onlow-carbon logistics is not much

In the process of studying the integrated cost of thetransport network and the optimal vehicle configurationresearchers incorporated the cost of risk management andthe psychological cost of waiting for rescue into the emer-gency reverse logistics model constructed the logisticsnetwork model and optimized the integrated cost oftransport network based on the trade-off of three conflictingobjectives logistics cost risk cost and psychological cost

2 Journal of Advanced Transportation

[19] ey improved the multiobjective particle swarmoptimization algorithm to solve the multiobjective enter-prises under the carbon tax policy [20] and improved thegenetic algorithm to optimize the transportation networkscost and distribution of the vehicle [21] By studying thesealgorithms this paper finally chooses to use the improvedparticle swarm optimization algorithm to solve thetransport network optimization problem between mul-tiple shippers and consignees Influenced by the relevantpolicies and environmental protection requirementsChinarsquos research on bulk cargo transport structural ad-justment has achieved certain results It has become aninevitable trend to shift the mode of transport of cargofrom road to rail transport However it can be seen fromthe above literature that researchers still have doubtsabout how to promote the ldquoroad to railrdquo of bulk cargo andreduce the integrated cost of the transportation industryFor the optimization of the low-carbon transport betweenmultiple shippers and multiple consignees the existingresearch needs to be further deepened To this end thispaper innovatively combines fourth-party logistics withbulk cargo green transport between multiple shippers andmultiple consignees rough the intelligent bulk cargofourth-party logistics network to attract more sources andreduce transportation cost improve the particle swarmoptimization algorithm to evaluate the fourth-party lo-gistics networkrsquos effect to prove the fourth-party logisticsrsquooptimization effect

3 Green Transport Mode of Bulk CargoesBased on Fourth-Party Logistics

According to the definition of the China Railway CustomerService Center bulk cargo is large cargo volume and stablecargo such as coal coke petroleum and metal ore whichcan be determined in advance for transportation needs Bulkcargo transport accounts for more than 50 of Chinarsquos cargotransport market ranking first among all types of freighttransport [3] e proportion of bulk cargo transport sharedby roads is up to 7506 [22] Due to the differences intransport means energy capacity and facilities rail trans-portrsquos energy consumption and emission intensities are only17 and 113 of road transport [22] Carbon emissions arehigher in road transport and lower in railway transport sothe implementation of ldquoroad to railrdquo to promote the transferof bulk cargo from road transport to rail transport hasbrought significant energy conservation and emission re-duction effects (unit turnover can save 86 of energyconsumption and reduce 92 of energy emissions [23])However railway bulk cargo transport is still challenging tocomplete ldquopoint-to-point door-to-doorrdquo transport etransport process is complicated and the transaction costsare high erefore it is necessary to promote the transfer ofbulk cargo transport from road transport to railwaytransport

Based on the existing research related to logisticsinformatization [24ndash28] this paper builds a green transportmode of bulk cargo through fourth-party logistics Com-pared with other researchersrsquo big data models this paper

incorporates carbon emission cost factors into the selectionand evaluation system of bulk cargo transport methods Itdetermines the optimal transport model by comparing theintegrated cost and carbon emissions between differenttransport modes within a limited time

31 Problem Description and Assumptions Compared withroad transport railway transport has advantages of low costlarge volume all-weather energy-saving and environmentalprotection safety and stability ere are also manyshortcomings such as complicated transport links high costsfor short-distance freight tight local transport capacitylimited timeliness and insufficient flexibility [29] In thispaper the cost of fourth-party logistics and carbon emis-sions is included in the green transport mode of bulk cargoesand the optimal transport method is selected by comparingdifferent transport costs and transport efficiencies

In the traditional bulk cargo transport network due toinsufficient informationization inflexibility inadequatepolicy support and imperfect multimodal transportmechanism bulk cargo rail transport is complicated and thetransaction procedures are numerous [30] e lack ofcommunication between shipper and cargo enterprise be-tween consignee and cargo enterprise and between shipperand consignee created a large amount of information gamecost which makes shipper more willing to choose roadrather than railway transport to bear the volume of bulkcargo As a result in promoting the green transportation ofbulk goods from road to railway a big data logistics servicethat can share transport network information simplifytransaction procedures and promote low-carbon emissionsis needed Fourth-party logistics can provide such logisticsservices

The use of fourth-party logistics to promote greentransport of bulk cargoes has the following effects (1) It isconducive to the application of intelligence informati-zation and big data (2) It simplifies the railway transportprocess and improves the service level of multimodaltransport with the railway as the core (3) It reduces theintegrated cost of green transport networks for bulkcargoes

Figure 1 shows the design of the fourth-party logisticsnetwork for bulk cargo In the fourth-party logistics net-work shippers release logistics orders through the fourth-party logistics e platform formulates transport plans aftercollecting detailed logistics information and distributes theinformation to the corresponding third-party logistics en-terprises through bidding Enterprises integrate the goods ofdifferent shippers through fourth-party logistics andtransport the integrated goods according to the require-ments of time destination and route In this transportprocess the enterprise enjoys certain autonomy It mustfollow the relevant transport requirements and time limitrequirements while undertaking logistics orders and is re-sponsible for the transport process When the order reachesthe designated consignee the driver will give feedback to theplatform and the task will end after confirmation by theplatform During this period all businesses will be

Journal of Advanced Transportation 3

completed through the fourth-party logistics platform Afterreceiving the relevant logistics task instructions eachmember should prepare in accordance with the transportplan formulated by the platform Each member can com-municate in real time and supervise the order processthrough the fourth-party logistics platform during thetransport process If there is any loss or accident in thetransport process the fourth-party logistics platform shallcoordinate by relevant rules

Figure 2 shows the schematic diagram of the fourth-party logistics network for bulk cargo e fourth-partylogistics is responsible for the bulk cargo delivery ofproducts between multiple supply and demand pointpairs of products e bulk cargo is transported to dif-ferent consignees i through n freight enterprises whereeach freight enterprise is responsible for providingtransport services to multiple supply and demand pointpairs e nodes in the network include shippers con-signees and freight enterprises Cargoes are sent fromshippers and transferred to individual consignees byfreight enterprises xijkn indicates the transport routefrom shipper j to consignee i In order to maximize theprofits of the overall logistics network it is necessary toreasonably plan the bulk cargo transport routes and thenode distribution locations to minimize the sum of thecosts of all the distribution routes xijkn It is crucial tocomprehensively consider the carbon emission and es-tablish the green transport network model of bulk cargoesbased on fourth-party logistics to minimize the integratedcost within a limited time

e carbon emission cost of the logistics networkinvolved in this paper mainly refers to the carbon dioxideemission cost generated by road or rail transport duringthe mainline transport process of the bulk cargo logisticsnetwork For the convenience of calculation this paperseparately sets the cost of carbon emission factor per unitdistance for road and rail transport α1 α2 [20] and cal-culates the overall carbon tax cost of logistics networkwithout detailed calculation of the carbon emissions of aspecific model In other words this paper takes the cost ofcarbon emission the starting cost of the service distri-bution of the cost of processing and transportation as thetotal cost and finds the integrated cost of the logisticsnetwork when the total is minimized is design avoidsthe problem of excessive objective functions in mostexisting low-carbon transport optimization studies It ismore suitable for the fourth-party logistics network thanthe single-objective optimization model

In order to simplify the calculation process this papermakes the following assumptions (1) e transport timeof the road in the transport network of the same con-dition is the same as that of the railway (2) e fourth-party logistics needs additional opening costs reducingthe transport networkrsquos carbon tax rate and informationcost after opening

32 Description of Model Components e relevant pa-rameters sets and decision variables involved in this modelare shown in Table 1

33 Model Building Based on the above symbol descrip-tions and variable definitions the integrated cost andcarbon emission of the fourth-party logistics network isshown as follows

MinC 1113944 yn H + CI( 1113857 + 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

Cnxijkn

+ 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

cijknxijkn d1jn + d

2in1113872 1113873 + α12E

(1)

E 1113944iisinV

1113944jisinV

1113944nisinV

1113944kisinK

β12qijknbpn d

1jn + d

2in1113872 1113873 (2)

e objective function formula (1) represents theoptimal integrated cost of the transport network in alimited time which includes the opening cost of theservices of freight enterprises loading and unloadingcosts of bulk cargo transport transport costs and thecarbon emission cost incurred by vehicles of freight en-terprises Formula (2) represents carbon emissions fromthe transport network

1113944jisinNcup J

1113944nisinN

1113944kisinK

xijkn bpn foralli isin Iforallp isin P (3)

1113944jisinNcup I

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK

xijkn 1 foralli isin Iforallp isin P (5)

1113944iisinN

1113944nisinN

1113944kisinK

xijkn 1 forallj isin J forallp isin P (6)

1113944jisinN

1113944nisinN

1113944kisinK

ap

ijkn 1 foralli isin Iforallp isin P (7)

1113944iisinN

1113944nisinN

1113944kisinK

ap


Constraints (3) and (4) indicate that if the freight en-terprise n of the logistics network is selected to serve thesupply point pair p there must be a transport mode con-nected to it to transport the cargo otherwise no freightenterprise or transport mode is selected Constraints (5) and(6) require that the bulk cargo be transported from shipper jto consignee i via freight enterprise n Constraints (7) and (8)require respectively that the shipper j and consignee i ofpoint of supply and demand point pair pmust transport theproducts via freight enterprise n


β12 1113944pisinP

bpn d

p leQnyn foralln isin N (9)

1113944pisinP

ap

ijkndp le qijknxijkn foralli isin Vforallj isin V forallk isin Kforalln isin N

(10)

β12 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

xijkn⎛⎝ ⎞⎠ 1113944

iisinI1113944jisinJ

1113944nisinN

1113944kisinK

ap

ijkn (11)

1113944iisinI

Si 1113944jisinJ

Dj (12)

Supplier

Online Bidding for Orders

Feedback Order

Feedback Order

Enterprise

Arrange Drivers Accordingto Relevant Requirements

Consignee

Fourth-party

Logistics

Assign

OrderAssignOrder

DriverVehicle

CargoTransport

Figure 1 Bulk cargo fourth-party logistics network design

Fourth-party Logistics Network

Railway Transport

Railway Transport

Railway Transport

Railway Transport

Road Transport

Road Transport

Road Transport

Road Transport

Enterprise 1

Enterprise 2

Enterprise 3

Enterprise n

Supplier 1

Supplier 2

Supplier 3

Supplier j

Consignee 1

Consignee 2

Consignee 3

Consignee 4

Consignee i

Figure 2 e fourth-party logistics network for bulk cargo


qijkn ge 0 foralli isin V forallj isin Vforallk isin K (13)

1113944jisinJ

qijkn le Si foralli isin I (14)

1113944iisinI

qijkn leDj forallj isin J (15)

1113944pisinP

bpn qijkn leWnyn foralln isin N (16)

Constraint (9) is the carbon emission constraint offreight enterprises Constraint (10) is a transport routecapacity constraint of freight enterprise n Constraint(11) aims to determine the carbon emissions of freightenterprises Constraint (12) ensures that the volume offreight sent by all shippers is the same as the consigneereceives Constraints (13)ndash(15) indicate the capacityconstraints for transport between points Constraint (16)represents the capacity constraint on the bulk cargo offreight enterprise n

ap

ijkn isin 0 1 foralli isin Vforallj isin Vforalln isin Vforallk isin Kforallp isin P

(17)

bpn isin 0 1 foralln isin N forallp isin P (18)

xijkn isin 0 1 foralli isin Vforallj isin V foralln isin Vforallk isin K (19)

yn isin 0 1 foralln isin N (20)

Constraints (17)ndash(20) are 0ndash1 decision variables

34 Algorithm Design Optimizing the green transportnetwork for bulk cargoes of fourth-party logistics is anextension of the traditional logistics transport problem Insolving theminimum cost of the logistics network due to theinvolvement of various shippers freight enterprises andlocations of shippers it is also an NP problem is paperinvolves multiple shippers and multiple consignees ecalculation scale is large and the solution is complicated

Table 1 Parameters sets and decision variables

Symbol Descriptiond1

jn e distance from the shipper j to the freight n (km)d2

in e distance from the consignee i to the freight n (km)

qijkne unit transportation capacity of the vehicle k from the shipper j to the consignee i through the

freight enterprise ncijkn e cost of per vehicle per unit distance (yuankm)I e set including suppliers indexed by i I 0 1 2 iJ e set including shippers indexed by j J 0 1 2 jP e set including supply and demand points from shipper j to consignee iK e set including freight vehicles indexed by kN e set including freight enterprises indexed by n N 0 1 2 nGn Maximum load of the freight enterprise nCn e cost of loading and unloading cargo of the freight enterprise n per unit timeSi e set including cargo quantity obtained by the consignee iDj e set including cargo quantity issued by the shipper jWn e freight handling capacity of the freight enterprise nα1 α2 e carbon tax rate of bulk cargo transport network withwithout fourth-party logisticsβ1 β2 e carbon emission factors per unit distance per product in roadrail transportdp e number of emergency supplies between the point pair p in the period p isin PV I cup J cup NH e opening cost of the fourth-party logistics serviceCI e information transaction cost of the traditional bulk cargo transport networkE Carbon emissions from bulk cargo transport networksQn Maximum load of carbon emission capacity of all vehicles in the freight enterprise n

xijknisin0 1If the vehicle k is transported from the shipper j to the consignee i via the freight enterprise n then

xijkn 1 otherwise xijkn 0ynisin0 1 If the fourth-party logistics service is opened then yn 1 otherwise yn 0

ap

ijkn isin 0 1 If the shipper j to the consignee i is transported by the vehicle k and serves the point pair p via freight

enterprise n then apijkn 1 otherwise apijkn 0b

pn isin 0 1 If the freight enterprise n is served the supply and demand point p then b

pn 1 otherwise b

pn 0

X+ xijkn|foralli isinV forallj isinV foralln isinVforallk isinKX- apijkn|foralli isinV forallj isinV foralln isinVforallk isinK forallp isin PY yn|foralln isinN


e corresponding intelligent optimization algorithm needsto be selected to solve the problem erefore this paperimproves the particle swarm optimization (PSO) algorithmto optimize the green transport of bulk cargoes based onfourth-party logistics

341 Particle Swarm Optimization Algorithm Particleswarm optimization (PSO) algorithm is an effectiveglobal optimization algorithm derived from the study ofbird predation behavior and is used to solve the opti-mization problem Similar to other evolutionary algo-rithms PSO also searches for optimal solutions incomplex spaces through cooperation and competitionamong individuals PSO has the characteristics of evo-lutionary computing and swarms intelligence It is anoptimization algorithm based on swarm intelligencetheory which guides the optimization search throughswarm intelligence generated by the competition andcooperation among particles in the swarm PSO retains apopulation-based global search strategy compared withtraditional evolutionary algorithms but its ldquospeed-dis-placementrdquo model is simple to operate and avoidscomplex genetic operations Its unique memory enables itto dynamically track the current search situation to adjustthe search strategy

In PSO each optimization problemrsquos solution is con-sidered a bird in the search space that is ldquoparticlerdquo First asthe initial population is generated a group of particles israndomly initialized in a feasible solution space Eachparticle is a feasible solution and an objective functionevaluates its fitness value Each particle moves in the so-lution space and its flight direction and distance are de-termined by velocity Usually the particle follows thecurrent optimal particle to search in the solution spaceDuring each iteration the particle will track two ldquoextremevaluesrdquo to update itself One is the optimal solution foundby itself and the other is the optimal solution currentlyfound by the whole population is extremum is theoptimal global solution

is paper assumes that the vehicle k transportedbetween shipper j and consignee i refers to particles eseparticles are searched in the p-dimensional space eswarm consisting of P particles is X X1 X2 XP wherethe position of each particle Xk Xk1 Xk2 XkP rep-resents a feasible solution to the optimization problemEach particle has a velocity which is Vk Vk1 Vk2VkP Particles search for new solutions by constantlyadjusting their positions and each particle can rememberthe optimal solution it has searched for In each iterationthe velocity and position of the particle are constantlyupdated according to the optimal historical position of theindividual particle Lk Lk1 Lk2 LkPT and the globaloptimal position Lg Lg1 Lg2 LgP1113966 1113967

Tto achieve

population purification e formula for updating thevelocity and position of the kth p-dimensional particle isas follows

Vz+1kp ω times V

zkp + c1 times r1 times P

zkp minus X

zkp1113872 1113873 + c2 times r2 times P

zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)

In formulas (21) and (22) z represents the currentnumber of iterations c1 c1 represent the acceleration factorr1 r2 represent the random number between [0 1] and ω isthe inertia weight and its size has a great effect on theconvergence of PSO

342 Improved Implementation of Particle Swarm Optimi-zation Algorithm is paper designs a constraint process-ing mechanism based on the traditional PSO algorithm thatallocates the region and location of the solution by Paretodominance It then searches through the feasible region bythe PSO algorithm After several iterative evolutions theinfeasible solution is finally eliminated and the optimalsolution is retained According to the well-defined particleswarm structure this paper adopts an improved particleswarm optimization algorithm (IPSO) to optimize themany-to-many transport cost of the fourth-party logisticsnetwork under low-carbon emissions

Step 1 Initialize all parameters such as Vk Lk and Lg in theparticle swarm and generate the initial Pareto solution setwith Vk Lk and Lg

Step 2 Calculate fitness with Vk Lk and Lg

Step 3 According to the Pareto solution set of differentshippers and consignees update Vk Lk and Lg

Step 4 Correct the position and velocity of the particlesbeyond the boundary For each particle k compare its fitnessvalue with the best position Lk in the Pareto solution set If itis better update Lk e update of Lg is the same as Lk

Step 5 e algorithm is ended if the end condition isreached (get a sufficiently good position or the maximumnumber of iterations) [20] e optimal global solution isoutput otherwise continue the iteration

e pseudocode of the IPSO algorithm is shown inAlgorithm 1

343 Design of Constraint Processing MechanismConstraints are an essential part of many-to-many logisticsnetwork optimization problems e key to using the IPSOalgorithm to solve the many-to-many optimization problemlies in designing effective constraint processing mechanismsBecause there are many constraints involved this paperdesigns a flow allocation algorithm a transportation capacitylimit conversion algorithm and a capacity adjustment al-gorithm to solve the mixed-integer programming modelbetween multiple shippers and multiple consignees underconstraints


In order to determine the cargo flow between the variousfacilities this paper converts formulas (3)ndash(8) (11) and (12)into algorithmic languages for expression under the con-dition of solving objective functions (1) and (2) to select themethod with the lowest integrated cost and carbon emissionamong the various nodes for bulk cargo transport In orderto ensure that the transportation volume between nodesdoes not exceed its carrying capacity this paper uses apenalty function to convert transportation capacity con-straints (9) and (10) to ensure that the bulk cargo volume ofeach transportation mode does not exceed its correspondingcapacity limit In order to make the opened nodes operatewithin their capabilities this paper converts capacity con-straints (13)ndash(16) into MATLAB language for expressionthus selecting the node with the lowest integrated cost foropening

4 Experiments and Data Analysis

In order to optimize the fourth-party green transport net-work mode of bulk cargoes this paper uses IPSO to test thefourth-party green transport network of bulk cargoes withthree different conditions large medium and small emodels and algorithms are coded inMATLAB language andexperiments are performed on different computers enodes and data involved in this experimental calculation aremainly based on Li et alrsquos improved particle swarm opti-mization algorithm and reasonable adjustment and opti-mizations [20]

Assume that in the fourth-party logistics mode dif-ferent freight enterprises undertake bulk cargo transporttasks from multiple shippers j to consignee i e carbonemission factor per unit product unit distance in roadtransport β1 is 0056 (kgtmiddotkm) and the carbon emissionfactor per unit product unit distance in railway transport

β2 is 0017 (kgtmiddotkm) [20] e carbon tax rate of transportnetworks with fourth-party logistics service α1 is 250yuant and that of the transport network without fourth-party logistics service α is 300 yuant [31] e cost ofloading and unloading the bulk cargo during the road-to-rail transport is 1000 yuan (combined with the openingcost of the rail transport service when calculating) andeach transport process includes two times of loading andunloading In the bulk cargo logistics network assumed inthis paper because the front-end transport and the ter-minal transport part of the rail or road transport processuse road transportation the costs incurred are the samethat is only the cost of railway transport in the mainlinetransport the sum of the loading and unloading costs atboth ends and the road transport integrated cost areconsidered to be compared in the calculation of the in-tegrated cost For simplicity this paper does not considerthe fixed construction costs of shippers consignees andfreight enterprises

41 Green Transport Optimization for the Small Bulk CargoTransport Network A small bulk cargo network consistsof 2 shippers 2 freight enterprises and 2 consigneesAfter receiving the bulk cargo provided by the shipperthe freight enterprises in the network can choose twomain transport modes road and rail transport to transferthe bulk cargo to the consignee and the shipper has threevehicles under each transport mode to perform the de-livery task e related operation and management dataof each node of the logistics network are shown in Table 2

e shipper contracts bulk cargo transport to thecorresponding freight enterprises and chooses the roadand rail transport modes to transport the cargo to theconsignee according to the actual situation e distancesbetween the nodes are shown in Table 3

(i) FOR each particle k(ii) Initialize Vk particle velocity(iii) Initialize Lk the historical optimal position(iv) Initialize Lg the optimal global position(v) Initialize Pareto solution set Generate Pareto solution set with Vk Lk and Lg

(vi) Initialize N population number(vii) Calculate fitness Calculate fitness with involving parameters(viii) FOR each particle k(ix) Update Vk according to the Pareto solution set(x) Update fitness update fitness with new Vk(xi) IF the fitness value is better than Lk in history(xii) Lk the fitness value(xiii) END IF(xiv) IF the fitness value is better than Lg in history(xv) Lg the fitness value(xvi) END IF(xvii) Update the new Pareto solution set Update the Pareto solution set with new Vk Lk and Lg

(xviii) END FOR(xix) i i+1(xx) WHILE maximum iterations or get a sufficiently good position

ALGORITHM 1 e improved particle swarm optimization algorithm


Combining the above data this paper assumes that thesupply and demand of each shipper and consignee and thecirculation of bulk cargoes are as shown in Table 4

After the MATLAB evolution algorithm the integratedcost of road and railway transport in the condition of thesmall bulk cargo transport network in a limited time isobtained as shown in Figure 3 In the condition of the smallbulk cargo transport network because the number ofshippers and consignees is less than the number of vehiclesheld by freight enterprises the logistics network transportroute is not unique When the fourth-party logistics serviceis not selected the integrated cost of the road transportnetwork is 187210 yuan and the integrated cost of therailway transport network is 127330 yuan when selectingthe fourth-party logistics service the integrated cost of theroad transport network is 198700 yuan and the integratedcost of the railway transport network is 139230 yuan Be-sides in the small bulk cargo transport network the carbonemissions from road transport are 102 tons and the carbonemissions from railway transport are 20 tons Promoting thetransfer of transport methods from road to railway canreduce carbon emissions by 8039

42GreenTransportOptimization for theMediumBulkCargoTransport Network A medium bulk cargo network consistsof 3 shippers 3 freight enterprises and 4 consignees After

receiving the bulk cargo provided by the shipper the freightenterprise in the network can choose two main transportmodes road and rail transport to transfer the bulk cargo tothe consignee e shipper has three vehicles under eachtransport mode to perform the delivery task e relatedoperation andmanagement data of each node of the logisticsnetwork are shown in Table 5

e shipper contracts bulk cargo transport to the cor-responding freight enterprises and chooses the road and railtransport modes to transport the cargo to the consigneeaccording to the actual situation e distances between thenodes are shown in Table 6


After the MATLAB evolution algorithm the integratedcost of road and railway transport of the medium bulk cargotransport network in a limited time is shown in Figure 4When the fourth-party logistics service is not selected theintegrated cost of the road transport network is 540260yuan and the integrated cost of the railway transport net-work is 366820 yuan when selecting the fourth-party lo-gistics service the integrated cost of the road transportnetwork is 546800 yuan and the integrated cost of therailway transport network is 374520 yuan Besides in themedium bulk cargo transport network the carbon emissionsfrom road transport are 292 tons and the carbon emissionsfrom railway transport are 60 tons Promoting the transferof transport methods from road to railway can reducecarbon emissions by 7945

Table 8 shows the route flow of different consigneesreceiving the products sent by the shipper in the form ofroad and rail transport Taking consignee 1 as an examplewhen choosing road transport the cargoes received aretransported by shipper 2 via the third freight vehicle of

Table 2 Parameter of the small bulk cargo transport network

Parameter description Parameter Value Parameter description Parameter Value

Transport capacity of each vehicle Qk 1000 t Number of vehicles in eachfreight enterprise K 3

Transport capacity of freight enterprise 1 G1 1800 t Cargo required by consignee 1 S1 800 tTransport capacity of freight enterprise 2 G2 1900 t Cargo required by consignee 2 S2 1000 tFreight volume from shipper 1 D1 3000 t Freight volume from shipper 2 D2 2850 t

Maximum load of road freight enterprise W1 2000 t Rail handing capacity of freightenterprise 1 Wrsquo

1 2500t

Maximum load of rail freight enterprise W2 2100 t Rail handing capacity of freightenterprise 2 Wrsquo

2 2600t

Information transaction cost of traditionaltransportation network CI 3000 yuan Opening cost of fourth-party

logistics service H 30000 yuan

Unit road transport costs of enterprises cijkn16593 yuan

ttimes kmUnit rail transport costs of

enterprises crsquoijkn11062 yuan

ttimes km

Table 3 e distance between nodes in the small bulk cargo transport network (km)

Freight enterprise 1 Freight enterprise 2Shipper 1 140 190Shipper 2 210 175Consignee 1 325 320Consignee 2 350 320

Table 4 Circulation of cargoes at supply and demand points in thesmall bulk cargo transport network (0 means there is no supply anddemand relationship)

Consignee 1 Consignee 2Shipper 1 0 600Shipper 2 500 0


Table 5 Parameter of the medium bulk cargo transport network



Transport capacity of freight enterprise 1 G1 1800 t Cargo required by consignee 1 S1 800 tTransport capacity of freight enterprise 2 G2 1900 t Cargo required by consignee 2 S2 1000 tTransport capacity of freight enterprise 3 G3 2000 t Cargo required by consignee 3 S3 650 tFreight volume from shipper 1 D1 3000 t Cargo required by consignee 4 S4 700 tFreight volume from shipper 2 D2 2850 t Freight volume from shipper 3 D3 3500 t

Road handing capacity of freight enterprise 1 W1 2000 t Rail handing capacity of freightenterprise 1 W1prime 2500 t





Unit road transport costs of enterprises cijkn16593 yuant

times kmUnit rail transport costs of

enterprises cijknprime 11062 yuant

times km

Table 6 e distance between nodes in the medium bulk cargo transport network (km)

Freight enterprise 1 Freight enterprise 2 Freight enterprise 3Shipper 1 140 190 65Shipper 2 210 175 125Shipper 3 190 240 105Consignee 1 325 320 370Consignee 2 350 320 335Consignee 3 345 320 340Consignee 4 355 320 365

22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)

Number of iterations0 20 40 60 80 100 120 140 160 180 200

Integrated of the small bulk cargo transport network

Small Rail with 4PLSmall Rail without 4PL

Small Road with 4PLSmall Road without 4PL

times105

Figure 3 e integrated cost of the small bulk cargo transport network


freight enterprise 2 and by shipper 3 via the first freightvehicle of freight enterprise 2 In the selection of railtransport the cargoes received by shipper 2 are thosetransported by shipper 2 via the first freight vehicle of freightenterprise 3 and by shipper 3 via the third freight vehicle offreight enterprise 3

43 Green Transport Optimization for the Large Bulk CargoTransportNetwork A large bulk cargo network consists of 5shippers 5 freight enterprises and 6 consignees After

receiving the bulk cargo provided by the shipper the freightenterprise in the network can choose two main transportmodes road and rail transport to transfer the bulk cargo tothe consignee and the shipper has three vehicles under eachtransport mode to perform the delivery task e relatedoperation andmanagement data of each node of the logisticsnetwork are shown in Table 9



After the MATLAB evolution algorithm the integratedcost of road and railway transport of the large bulk cargotransport network in a limited time is shown in Figure 5When the fourth-party logistics service is not selected theintegrated cost of the road transport network is 1226300yuan and the integrated cost of the railway transport net-work is 824180 yuan When selecting the fourth-party lo-gistics service the integrated cost of the road transportnetwork is 1203100 yuan and the integrated cost of the

6

55

Inte

grat

ed co

st (y

uan)

Number of iterations

5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200

Integrated of the medium bulk cargo transport network

Medium Rail with 4PLMedium Rail without 4PL

Medium Road with 4PLMedium Road without 4PL

Figure 4 e integrated cost of the medium bulk cargo transport network

Table 7 Circulation of cargoes at supply and demand points in the medium bulk cargo transport network

Consignee 1 Consignee 2 Consignee 3 Consignee 4Shipper 1 0 600 0 400Shipper 2 500 0 200 300Shipper 3 300 400 450 0

Table 8e flow of supply and demand points in the medium bulkcargo transport network

Road transport route Rail transport routeConsignee Route flow Consignee Route flow1 2⟶2⟶1 (3) 1 2⟶3⟶1 (3)1 3⟶2⟶1 (3) 1 3⟶3⟶1 (3)2 1⟶2⟶2 (2) 2 1⟶3⟶2 (1)2 3⟶3⟶2 (2) 2 3⟶3⟶2 (3)3 2⟶3⟶3 (2) 3 2⟶2⟶3 (2)3 3⟶3⟶3 (1) 3 3⟶3⟶3 (1)4 1⟶2⟶4 (2) 4 1⟶2⟶4 (3)4 2⟶2⟶4 (2) 4 2⟶2⟶4 (1)


railway transport network is 808520 yuan Besides in themedium bulk cargo transport network the carbon emissionsfrom road transport are 640 tons and the carbon emissionsfrom railway transport are 132 tons Promoting the transferof transport methods from road to railway can reducecarbon emissions by 7938

Table 12 shows the route flow of different consigneesreceiving the products sent by the shipper in the form ofroad and rail transport

44 Algorithm Performance Analysis e algorithm isimplemented using MATLAB R2018a programming In orderto verify the algorithm performance this paper defines therelevant performance parameters of the algorithm e

algorithm runs 100 times BESTrepresents the best solution of100 BAD the worst solution MEAN represents the averagevalue and TIME represents the average time consumed by thealgorithm once We solve large medium and small railtransport networks using traditional PSO and IPSO respec-tively e results of the algorithm comparison are shown inTable 13

It shows that both PSO and IPSO can more effectivelysolve the problems for the three cases of different scales Forsmall logistics networks both algorithms obtain better so-lutions quickly However with the increase of the logisticsnetwork the IPSO has gradually shown higher stability thanthe traditional PSO By comparing the logistics networks ofthree scales it can be found that IPSO consumes less timeand has higher stability of the solution than PSO It means

Table 9 Parameter of the large bulk cargo transport network



Transport capacity of freight enterprise 1 G1 1800 t Cargo required by consignee 1 S1 800 tTransport capacity of freight enterprise 2 G2 1900 t Cargo required by consignee 2 S2 1000 tTransport capacity of freight enterprise 3 G3 2000 t Cargo required by consignee 3 S3 650 tTransport capacity of freight enterprise 4 G4 2100 t Cargo required by consignee 4 S4 700 tTransport capacity of freight enterprise 5 G5 1700 t Cargo required by consignee 5 S5 1100 tFreight volume from shipper 1 D1 3000 t Cargo required by consignee 6 S6 1000 tFreight volume from shipper 2 D2 2850 t Freight volume from shipper 4 D4 3200 tFreight volume from shipper 3 D3 3500 t Freight volume from shipper 5 D5 2700 t











times km

Table 10 e distance between nodes in the large bulk cargo transport network (km)

Freight enterprise 1 Freight enterprise 2 Freight enterprise 3 Freight enterprise 4 Freight enterprise 5Shipper 1 140 190 65 220 90Shipper 2 210 175 125 150 185Shipper 3 190 240 105 70 165Shipper 4 145 165 90 95 180Shipper 5 155 75 130 100 195Consignee 1 325 320 370 310 370Consignee 2 350 320 335 315 375Consignee 3 345 320 340 330 370Consignee 4 355 320 365 325 385Consignee 5 330 320 380 305 390Consignee 6 300 320 365 300 395


that compared with PSO IPSO can solve the proposedproblem faster and more accurately showing the improvedmethodrsquos effectiveness

45 Comparison of Results By comparing the costs of largemedium and small bulk cargo logistics networks under roadand rail transport conditions this paper investigates theeffect of low-carbon emission reduction effect of railwaytransport on road transport under different conditions oflow-carbon emissions e results are shown in Table 14

From the above in the small medium and large bulkcargo transport networks the integrated cost of rail trans-port is lower than that of road transport which plays the roleof cost optimization From the perspective of the benefits ofthe transport network with the continuous expansion of thebulk cargo transport networkrsquos scale the cost optimizationrate of the transfer of bulk cargo from road transport to railtransport has gradually increased From the perspective ofthe fourth-party logistics when the bulk cargo transportnetwork is small the integrated cost of the transport networkusing fourth-party logistics is more than the integrated cost

Table 11 Circulation of cargoes at supply and demand points in the large bulk cargo transport network

Algorithm Consignee 1 Consignee 2 Consignee 3 Consignee 4 Consignee 5 Consignee 6Shipper 1 0 600 0 400 0 300Shipper 2 500 0 200 300 250 0Shipper 3 300 400 450 0 0 400Shipper 4 400 0 0 300 450 500Shipper 5 0 0 350 450 500 0

Table 12 e flow of supply and demand points in the large bulkcargo transport network

Road transport route Rail transport routeConsignee Route flow Consignee Route flow1 2⟶1⟶1 (2) 1 2⟶2⟶1 (1)1 3⟶1⟶1 (3) 1 3⟶4⟶1 (3)1 4⟶2⟶1 (3) 1 4⟶1⟶1 (2)2 1⟶2⟶2 (3) 2 1⟶4⟶2 (2)2 3⟶4⟶2 (1) 2 3⟶1⟶2 (3)3 2⟶4⟶3 (2) 3 2⟶4⟶3 (2)3 3⟶2⟶3 (3) 3 3⟶2⟶3 (1)3 5⟶3⟶3 (3) 3 5⟶4⟶3 (2)4 1⟶4⟶4 (2) 4 1⟶2⟶4 (3)4 2⟶4⟶4 (1) 4 2⟶4⟶4 (3)4 4⟶2⟶4 (2) 4 4⟶1⟶4 (1)4 5⟶1⟶4 (1) 4 5⟶1⟶4 (3)5 2⟶4⟶5 (2) 5 2⟶2⟶5 (1)5 4⟶1⟶5 (1) 5 4⟶4⟶5 (3)5 5⟶3⟶5 (1) 5 5⟶1⟶5 (1)6 1⟶3⟶6 (2) 6 1⟶2⟶6 (1)6 3⟶4⟶6 (1) 6 3⟶2⟶6 (2)6 4⟶3⟶6 (3) 6 4⟶4⟶6 (1)

0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)


Integrated of the large bulk cargo transport network

Large Rail with 4PLLarge Rail without 4PL

Large Road with 4PLLarge Road without 4PL

Figure 5 e integrated cost of the large bulk cargo transport network


of the transport network not using fourth-party logisticsWith continuous expansion the cost optimization benefitsbrought by fourth-party logistics gradually increase sur-passing the transport network that does not use fourth-partylogistics From the perspective of low-carbon emissions invarious transport networks with the same conditions thecarbon emissions generated by the implementation ofrailway transport are reduced by about 80 compared to thecarbon emissions generated by road transport which showsthat the promotion of bulk cargoes from roads to railwaystransfer can play an essential role in reducing costs andemissions

In summary combining fourth-party logistics with greentransport of bulk cargoes is more suitable for more extensivetransport networks and transferring bulk cargo transport fromroad to railway It is possible to significantly reduce the in-tegrated cost and carbon emissions of bulk cargo transportnetworksis shows that with the continuous expansion of thelogistics network the optimization benefits of rail transport forthe optimization of integrated costs continue to increase andthe energy-saving and emission reduction effects are very re-markable Using the fourth-party logisticsmode to promote theldquoroad to railrdquo of bulk cargo under low-carbon emission con-ditions can sufficiently reduce the integrated cost of the logisticsnetwork

5 Conclusions and Discussion

At present Chinarsquos economy has changed from a high-speed growth stage to a high-quality development stage

and the peoplersquos attention to the sustainable developmentof the transportation industry has gradually increasedWith the changes in national policies and the adjustmentof the transport structure railways should actively in-crease their transportation capacity to reduce carbonemissions from the transportation industry and promotegreen transport

is paper discusses the optimization of bulk cargogreen transport through fourth-party logistics In theprocess of combining the rail transport of bulk cargoeswith fourth-party logistics this paper draws the followingconclusions (1) e fourth-party logistics guides thetransfer of cargo from road transport to rail transportwhich can effectively reduce carbon emissions andtransport costs (2) Fourth-party logistics is more suitablefor large-scale bulk cargo transport networks and haslower benefits for small-scale bulk cargo transport net-works (3) e particle swarm optimization algorithm isimproved It can be found that IPSO consumes less timeand has higher stability of the solution than PSO

is paper mainly considers the transport mode selectionenergy saving and emission reduction and cost optimizationbenefits of the fourth-party logistics for the green transport ofbulk cargoes However the optimization of the ldquofirst and lastkilometrerdquo of rail transport under the fourth-party logistics hasnot been further solved In addition this paper does not in-corporate reverse logistics into the bulk cargo green transportnetwork In future research this paper will further optimize themodel and theory on the one hand to solve the ldquofirst and lastkilometrerdquo problem in the fourth-party logistics model of bulk

Table 13 PSO and IPSO performance in bulk cargo transport networks of different sizes

Network scale Algorithm BEST BAD MEAN TIME

Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656

Table 14 Effect in bulk cargo transport networks of different sizes

Networkscale

4PLservice

Transportmode

Integrated cost(yuan)

Cost optimization rate()

Carbon emissions(ton)

Carbon emission optimizationrate ()

SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100

MediumNo Road 540260 3210 2920

7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660


cargo and on the other hand to explore the optimization ofmany-to-many reverse logistics under the fourth-partytransport network by adding reverse logistics on the basis of theexisting model

Data Availability

e data that support the findings of this study are availablefrom the corresponding author upon reasonable request

Conflicts of Interest

e authors declare that they have no conflicts of interest

Acknowledgments

e authors would like to express their gratitude to Edit-Springs (httpswwweditspringscom) for expert linguisticservices e authors would also like to pay tribute to MrWang Xingpeng Ms Zhang Pei and Ms Zhang Qinghongwhose profound knowledge and effort helped in revising thispaper is research was funded by the Hebei ProvincialDepartment of Education (grant no SD 2021040) andsupported by the Development Fund of Innovative ResearchTeam of Shijiazhuang Tiedao University (grant no SCT202003)

References

[1] J Wang ldquoe Establishment of International EmissionTrading Mechanism Under e Paris Agreement ProcessChallenge and its Implication for Chinardquo EnvironmentalProtection vol 13 pp 58ndash62 2021

[2] National Bureau of Statistics of China China StatisticalYearbook China Statistics Press Beijing China 2000-2016

[3] Ministry of Transport of China Statistics Bulletin on theDevelopment of the Transportation Industry 2000-2019

[4] General Office of the State Council e ree-Year ActionPlan for Promoting the Adjustment of the Transport Structure(2018-2020) 2018 httpwwwgovcnzhengcecontent2018-1009content_5328817htm

[5] National Development and Reform Commission GuidingOpinions on Accelerating the Construction of Railway SpecialLines 2019 httpswwwndrcgovcnfggzzcssfzzcgh201909t20190918_1195138htmlcode=ampstate=123

[6] K Nakamura and Y Hayashi ldquoStrategies and instruments forlow-carbon urban transport an international review ontrends and effectsrdquo Transport Policy vol 29 pp 264ndash2742013

[7] R Zhang and Q Liu ldquoCO2 emission minimizing for the time-dependent VRP in urban Areardquo Industrial Engineering ampManagement vol 20 pp 29ndash34 2015

[8] L Yang Y Cai X Zhong Y Shi and Z Zhang ldquoA carbonemission evaluation for an integrated logistics system-A casestudy of the port of Shenzhenrdquo Sustainability vol 9 no 3p 462 2017

[9] Y Zhou W Fang M Li and W Liu ldquoExploring the impactsof a low-carbon policy instrument a case of carbon tax ontransportation in Chinardquo Resources Conservation andRecycling vol 139 pp 307ndash314 2018 httpsdoiorg101016jresconrec201808015

[10] R Gedik H Medal C Rainwater E A Pohl and S J MasonldquoVulnerability assessment and re-routing of freight trains

under disruptions a coal supply chain network applicationrdquoTransportation Research Part E Logistics and TransportationReview vol 71 pp 45ndash57 2014 httpsdoiorg101016jtre201406017

[11] L Zhao Y Zhao Q Hu H Li and J Stoeter ldquoEvaluation ofconsolidation center cargo capacity and loctions for Chinarailway expressrdquo Transportation Research Part E Logistics andTransportation Review vol 117 pp 58ndash81 2018 httpsdoiorg101016jtre201709007

[12] J Guo J Wang Q Li and B Guo ldquoConstruction of pre-diction model of neural network railway bulk cargo floatingprice based on random forest regression algorithmrdquo NeuralComputing amp Applications vol 31 no 12 pp 8139ndash81452019 httpsdoiorg101007s00521-018-3903-5

[13] D Marchetti and W Peter ldquoEfficiency of the rail sections inBrazilian railway system using TOPSIS and a genetic algo-rithm to analyse optimized scenariosrdquo Transportation Re-search Part E Logistics and Transportation Review vol 1352020 httpsdoiorg101016jtre2020101858 Article ID101858

[14] H A Von der Gracht and I-L Darkow ldquoEnergy-constrainedand low-carbon scenarios for the transportation and logisticsindustryrdquo International Journal of Logistics Managementvol 27 pp 142ndash166 2013 httpsdoiorg101108IJLM-12-2013-0150

[15] S Turki S Didukh C Sauvey and N Rezg ldquoOptimizationand analysis of a manufacturing-remanufacturing-transport-warehousing system within a closed-loop supply chainrdquoSustainability vol 9 no 4 p 561 2017 httpsdoiorg103390su9040561

[16] W Yang L Cheng and L Li ldquoStudy on the influence of taxrestructuring on the profit model of 4th-party logisticsplatformsrdquo China Business and Market vol 30 pp 100ndash1072016

[17] W Zhang ldquoe Development Trend of Chinarsquos LogisticsIndustry in the Era of Internet Plusrdquo Technoeconomics ampManagement Research vol 4 pp 104ndash108 2018

[18] P Wang A Study of Information Service Paradigm of Gov-ernment Procurement Cloud Based on Value Chain Collabo-ration Tianjin University Tianjin China 2016

[19] Z-H Hu J-B Sheu and J-B Sheu ldquoPost-disaster debrisreverse logistics management under psychological costminimizationrdquo Transportation Research Part B Methodo-logical vol 55 pp 118ndash141 2013 httpsdoiorg101016jtrb201305010

[20] L I Li F Cheng X Cheng G Wang and T Pan ldquoEnterpriseremanufacturing logistics network optimization based onmodified multi-objective particle swarm optimization algo-rithmrdquo Computer Integrated Manufacturing Systems vol 24pp 2122ndash2132 2018

[21] W Hai-rui N-N Jia and Z Lu-ping ldquoDry-port-basedsustainable logistics network for Inland Provinces inMaritimeSilk roadrdquo Systems Engineering vol 37 pp 63ndash73 2019

[22] Z Gu ldquoAn analysis on the demand characteristics of railwaybulk goods transportrdquo Railway Transport and Economyvol 41 pp 78ndash83 2019

[23] X Tao Q Wu and C Yin ldquoCO2 reduction potential ofshifting containers from road to rail for green transportdevelopmentrdquo Advances in Climate Change Research vol 11pp 1ndash9 2019

[24] M De Gennaro P Elena and G Martini ldquoBig data forsupporting low-carbon road transport Policiesin Europeapplicationrdquo Big Data Research vol 15 pp 11ndash25 2019


[25] E Menezes A Gori Maia A G Maia and C S de CarvalholdquoEffectiveness of low-carbon development strategies evalu-ation of policy scenarios for the urban transport sector in aBrazilian megacityrdquo Technological Forecasting and SocialChange vol 114 pp 226ndash241 2017 httpsdoiorg101016jtechfore201608016

[26] T Kawasaki and T Matsuda ldquoContainerization of bulktrades a case study of US-Asia wood pulp transportrdquo Mar-itime Economics amp Logistics vol 17 no 2 pp 179ndash197 2015httpsdoiorg101057mel201416

[27] J Wu and Y Wang ldquoDistribution of the emergency suppliesin the COVID-19 pandemic a cloud computing based ap-proachrdquo Mathematical Problems in Engineering vol 2021pp 1ndash18 2021 httpsdoiorg10115520215972747 ArticleID 5972747

[28] J Yang and W Wan ldquoCarbon emissions and reductionscenarios of transportation in Jiangsu Provincerdquo Highwayvol 11 pp 155ndash159 2017

[29] Y Ding ldquoRational Division of Labor and Coordinated De-velopment of Railway andHighway Logistics TransportationrdquoRailway Transport and Economy vol 38 pp 21ndash25 2016

[30] X Qi T He and B Mao ldquoe development and currentStatus of railway container intermodal transport in ChinardquoJournal of Transportation Systems Engineering and Informa-tion Technology vol 18 pp 194ndash200 2018

[31] Q HeW Luo and L Li ldquoLogistics network planning methodresearch under carbon tax constraint and coexist transportrdquoJournal of Railway Science and Engineering vol 11 pp 131ndash136 2014


to the distribution problem of ldquoroad to railrdquo [5] In additionthe State Railway Administration of China has taken anumber of measures in railway logistics infrastructurerailway transport prices and preferential transport policiesto deepen the reform of railway freight transport and ac-celerate the process of green transport

Recently the traditional bulk cargo transportation in-dustry has gotten rid of the low-price-first model Shippersrsquodemand for the safety convenience and delivery time oftransport is increasingly higher e transportation industryurgently needs a new model to improve the economicbenefit In order to optimize transport costs under low-carbon emissions the following innovations are made in thispaper (1) e fourth-party logistics promotes the rapidtransfer of bulk cargo from road transport to railwaytransport improves transport efficiency constructs an in-telligent transportation system and optimizes the integratedcost of bulk cargo transport (2) Unlike traditional bulkcargo transport this paper focuses on the transport processbetween multiple shippers and freight enterprises More-over the improved particle swarm optimization algorithmcompares the integrated cost and carbon emissions of roadand rail transport under different conditions within a limitedtime

e remainder of the paper is structured in the followingmanner Section 2 presents an overview of the relevantliterature focusing on green transport bulk cargo transportfourth-party logistics and algorithm design Section 3 buildsthe optimization model of bulk cargo green transport modebased on fourth-party logistics while Section 4 uses differentexperiments to illustrate the optimization effect of thefourth-party logistics of bulk cargoes on green transportSection 5 summarizes the findings and outlines potentialfuture research extensions

2 Literature Review

In order to promote the development of green transportresearchers have carried out much research on the low-carbon emission reduction of the transportation industryey analyzed the local transportation system through theCUTE matrix and pointed out that the implementation oflow-carbon policies and improvement of transportationtechnology are essential means to reduce the carbonemissions of the transportation industry [6] A model ofurban vehicle distribution with low carbon energy savingand low cost was constructed and genetic algorithms weredesigned to optimize vehicle distribution routes [7] Mea-sured the carbon emissions of the integrated logistics systemof the port and proved that the carbon emissions oftransport heavy equipment and logistics services accountfor a significant proportion of the transportation system [8]ey focused on the impact of carbon taxes on differenttransport modes and compared the carbon emissions andeconomic benefits of Chinarsquos railways roads and waterwaysthrough a general equilibrium model [9] e above re-searchersrsquo research on low-carbon emissions provides animportant idea for this article to explore green logisticsHowever there are still not many studies on promoting

green logistics in academia and it needs to be furtherimproved

In analyzing the characteristics of bulk cargo and pro-moting the green transport of bulk cargo researchers basedon the case of coal railway transport networks used bi-leveloptimization to explore the route of railway freight under thecondition of network interference [10] ey used a complexnetwork to evaluate the freight capacity of Chinarsquos road andrailway nodes and explored the differences in the trans-portation of bulk cargoes by road and railway [11] Combinethe bulk cargo price prediction model with neural networkalgorithms to improve railway freight volume predictionspeed and accuracy [12] ey also combined the TOPSISmethod to analyze the efficiency of bulk cargo transport inthe Brazilian railway system proposed some methods foroptimizing the rail transport of bulk cargoes and improvedrail freight efficiency [13] However researchersrsquo research onbulk cargoes mostly focuses on water transport and theresearch on rail transport of bulk cargoes needs furtherinnovation

In the research of transport informatization researcherscombined the views of 48 experts from 20 countries toexplore the supply chain management model of the futuretransportation industry and logistics industry from the fiveDelphi methods of energy emissions consumer behaviorfuture transportation mode future supply design and in-novation and concluded that a supply chain managementmodel that can help decision making information sharingand sustainability would emerge in the future and thefourth-party logistics could provide this management model[14] ey proposed a discrete flow model to describe thereverse logistics supply chain of remanufactured productswhich provided some ideas for the research of fourth-partylogistics [15] Constructed an ldquoN + 1 + Nrdquo fourth-partylogistics model based on Chinarsquos ldquoVAT Reformrdquo policy toachieve overall symmetry in logistics business informationand capital flows and explored the fourth-party logisticsprofit model [16] Combined with Internet technology itintegrated financial institutions consulting companies andother industries into the logistics industry chain to promotethe development of industry informatization and improvethe efficiency of logistics industry operation [17] Re-searchers integrated the government procurement infor-mation network through cloud technology to improvegovernment procurement efficiency and optimise supplychain management [18] e above researchersrsquo research onthe combination of logistics and big data provides manyideas for the research of fourth-party logistics in this paperHowever the existing logistics informatization researchmostly focuses on emergency logistics and the research onlow-carbon logistics is not much

In the process of studying the integrated cost of thetransport network and the optimal vehicle configurationresearchers incorporated the cost of risk management andthe psychological cost of waiting for rescue into the emer-gency reverse logistics model constructed the logisticsnetwork model and optimized the integrated cost oftransport network based on the trade-off of three conflictingobjectives logistics cost risk cost and psychological cost



















1113944jisinV

1113944nisinN

1113944kisinK

Cnxijkn

+ 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

cijknxijkn d1jn + d

2in1113872 1113873 + α12E

(1)

E 1113944iisinV

1113944jisinV

1113944nisinV

1113944kisinK

β12qijknbpn d

1jn + d

2in1113872 1113873 (2)


1113944jisinNcup J

1113944nisinN

1113944kisinK


1113944jisinNcup I

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK


1113944iisinN

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK

ap


1113944iisinN

1113944nisinN

1113944kisinK

ap




β12 1113944pisinP

bpn d


1113944pisinP

ap


(10)

β12 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

xijkn⎛⎝ ⎞⎠ 1113944

iisinI1113944jisinJ

1113944nisinN

1113944kisinK

ap

ijkn (11)

1113944iisinI

Si 1113944jisinJ

Dj (12)

Supplier


Feedback Order

Feedback Order

Enterprise


Consignee

Fourth-party

Logistics

Assign

OrderAssignOrder

DriverVehicle

CargoTransport



Railway Transport

Railway Transport

Railway Transport

Railway Transport

Road Transport

Road Transport

Road Transport

Road Transport

Enterprise 1

Enterprise 2

Enterprise 3

Enterprise n

Supplier 1

Supplier 2

Supplier 3

Supplier j

Consignee 1

Consignee 2

Consignee 3

Consignee 4

Consignee i




1113944jisinJ


1113944iisinI


1113944pisinP



ap


(17)














ap




pn 1 otherwise b

pn 0







Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References




















































1113944jisinV

1113944nisinN

1113944kisinK

Cnxijkn

+ 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

cijknxijkn d1jn + d

2in1113872 1113873 + α12E

(1)

E 1113944iisinV

1113944jisinV

1113944nisinV

1113944kisinK

β12qijknbpn d

1jn + d

2in1113872 1113873 (2)


1113944jisinNcup J

1113944nisinN

1113944kisinK


1113944jisinNcup I

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK


1113944iisinN

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK

ap


1113944iisinN

1113944nisinN

1113944kisinK

ap




β12 1113944pisinP

bpn d


1113944pisinP

ap


(10)

β12 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

xijkn⎛⎝ ⎞⎠ 1113944

iisinI1113944jisinJ

1113944nisinN

1113944kisinK

ap

ijkn (11)

1113944iisinI

Si 1113944jisinJ

Dj (12)

Supplier


Feedback Order

Feedback Order

Enterprise


Consignee

Fourth-party

Logistics

Assign

OrderAssignOrder

DriverVehicle

CargoTransport



Railway Transport

Railway Transport

Railway Transport

Railway Transport

Road Transport

Road Transport

Road Transport

Road Transport

Enterprise 1

Enterprise 2

Enterprise 3

Enterprise n

Supplier 1

Supplier 2

Supplier 3

Supplier j

Consignee 1

Consignee 2

Consignee 3

Consignee 4

Consignee i




1113944jisinJ


1113944iisinI


1113944pisinP



ap


(17)














ap




pn 1 otherwise b

pn 0







Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References










































1113944jisinV

1113944nisinN

1113944kisinK

Cnxijkn

+ 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

cijknxijkn d1jn + d

2in1113872 1113873 + α12E

(1)

E 1113944iisinV

1113944jisinV

1113944nisinV

1113944kisinK

β12qijknbpn d

1jn + d

2in1113872 1113873 (2)


1113944jisinNcup J

1113944nisinN

1113944kisinK


1113944jisinNcup I

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK


1113944iisinN

1113944nisinN

1113944kisinK


1113944jisinN

1113944nisinN

1113944kisinK

ap


1113944iisinN

1113944nisinN

1113944kisinK

ap




β12 1113944pisinP

bpn d


1113944pisinP

ap


(10)

β12 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

xijkn⎛⎝ ⎞⎠ 1113944

iisinI1113944jisinJ

1113944nisinN

1113944kisinK

ap

ijkn (11)

1113944iisinI

Si 1113944jisinJ

Dj (12)

Supplier


Feedback Order

Feedback Order

Enterprise


Consignee

Fourth-party

Logistics

Assign

OrderAssignOrder

DriverVehicle

CargoTransport



Railway Transport

Railway Transport

Railway Transport

Railway Transport

Road Transport

Road Transport

Road Transport

Road Transport

Enterprise 1

Enterprise 2

Enterprise 3

Enterprise n

Supplier 1

Supplier 2

Supplier 3

Supplier j

Consignee 1

Consignee 2

Consignee 3

Consignee 4

Consignee i




1113944jisinJ


1113944iisinI


1113944pisinP



ap


(17)














ap




pn 1 otherwise b

pn 0







Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References



































β12 1113944pisinP

bpn d


1113944pisinP

ap


(10)

β12 1113944iisinV

1113944jisinV

1113944nisinN

1113944kisinK

xijkn⎛⎝ ⎞⎠ 1113944

iisinI1113944jisinJ

1113944nisinN

1113944kisinK

ap

ijkn (11)

1113944iisinI

Si 1113944jisinJ

Dj (12)

Supplier


Feedback Order

Feedback Order

Enterprise


Consignee

Fourth-party

Logistics

Assign

OrderAssignOrder

DriverVehicle

CargoTransport



Railway Transport

Railway Transport

Railway Transport

Railway Transport

Road Transport

Road Transport

Road Transport

Road Transport

Enterprise 1

Enterprise 2

Enterprise 3

Enterprise n

Supplier 1

Supplier 2

Supplier 3

Supplier j

Consignee 1

Consignee 2

Consignee 3

Consignee 4

Consignee i




1113944jisinJ


1113944iisinI


1113944pisinP



ap


(17)














ap




pn 1 otherwise b

pn 0







Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References




































1113944jisinJ


1113944iisinI


1113944pisinP



ap


(17)














ap




pn 1 otherwise b

pn 0







Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References







































Tto achieve


Vz+1kp ω times V


zkp minus X


zkp minus X

zkp1113872 1113873

(21)

Xz+1kp X

zkp + V

z+1kp (22)




































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References




























































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References
















































1 2500t


2 2600t






ttimes km


















times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References















































times km



22

21

2

19

18

17

16

15

14

13

12

Inte

grat

ed co

st (y

uan)





times105









6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References









































6

55

Inte

grat

ed co

st (y

uan)


5

45

4

35

times105

0 20 40 60 80 100 120 140 160 180 200





























times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References






















































times km











0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References










































0 20 40 60 80 100 120 140 160 180 200

125

115

11

105

1

095

09

085

08

12

times106

Inte

grat

ed co

st (y

uan)
















Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References












































Small PSO 127330 127330 127330 90721IPSO 127330 127330 127330 72899

Medium PSO 366820 372820 366880 130755IPSO 366820 366820 366820 122844

Large PSO 824180 837180 824960 212474IPSO 824180 837180 824440 201656


Networkscale

4PLservice

Transportmode





SmallNo Road 187210 3196 1020

8039Railway 127330 200

Yes Road 198700 2993 510Railway 139230 100


7945Railway 366820 600

Yes Road 546800 3151 1460Railway 374520 300

LargeNo Road 1226300 3279 6400

7938Railway 824180 1320

Yes Road 1203100 3280 3200Railway 808520 660



Data Availability




Acknowledgments


References




































Data Availability




Acknowledgments


References











































Documents

Research on Green Transport Mode of Chinese Bulk Cargo