Table1:Meanandstandarddeviationofthesamplesclassifiedwithfactorsaffectingthe decisiont o usePurpleLineMRTservice PurpleLineMRTprovidesfastservicewithoutwaitinglong PurpleLineMRTprovidess
Trang 1FactorsaffectingPassengers’DecisionstoUsetheMetropolita
n RapidTransitChalongRatchadhamLine(PurpleLine)
KrongthongHeebkhoksung
BansomdejchaoprayaRajabhatUniversity
Abstract
Theobjectivesofthisresearchweretostudyfactorsaffectingpassengers’decisionstousetheMetropolitanR
a p i d TransitChalongRatchadhamLine(PurpleLine)andtostudywhetherpassengersdecidedtouseMR
TC h a l o n g RatchadhamLine.ThesettingofthisstudywasareaaroundMRTChalongRatchadhamLine(Pur pleLine).Dataw e r e collectedi n M a r c h 2 0 1 7 T h e a i m ofthisr e s e a r c h w a s tos t u d y p r e d i
c t o r f a c t o r s a n d p a s s e n g e r s ’ behaviorsinfluencingthepossibilityofusingPurpleLineMR T.Mathematicalmodelscouldbed e v e l o p e d topredictthedecisiontousePurpleLineMRT.Thepred ictiveresultswereusedtoplantransportmanagement.LogisticRegressionAnalysiswasusedtodetermine therelationship.Theresultsofstudyingtherelationships ho w e d thatp r e d i c t o r f ac to rs i nfl ue nc i ng t hepo s s i bi l i t y ofus i ngP ur pl e Li ne MRTi nc lude d a v a i l a b l e r o u t e s ,
f a s t , safe,andconvenienttransportaswellasotherfactors.BycomparingPurpleLineMRTusage,thep assengers’decisiontousethislinewasmorethantheirdecisionnottousetheservice
Keywords:MetropolitanRapidTransitChalongRatchadhamLine,PurpleLine
1.Introduction
Thep o l i c y ofdecentralizingthep r o s p e r i t y tol o c a l administrationa r e a s h a s beenformulat edbythegovernmentmeanwhilenumberofpopulationinthemetropolitanareaisstillincreasing.Thepric eoftypicalr e s i d e n c e s inBangkokisexpensive.Mostpeopleareunabletofindaresidenceinthe cityandareforcedtolivei n them e t r o p o l i t a n area.However,m o s t e m p l o y m e n t sourcesares
t i l l locatedi n B a n g k o k a n d a r e increasing.Thegreaternumberofpeoplelivinginouterareasneedsfo rstayinginBangkok,resultingintrafficjam.MetropolitanRapidTransitsystemisimportanttofacilitat ethepublictransportwithouttheusageofprivatecar.Presently,MRTstructureisconstructedinBan gkokandMetropolitanArea.TheMetropolitanR a p i d TransitChalongRatchadhamLine(Purpl eLine)containstotaldistanceof23km.and16stations.ThebehaviorsofusingPurpleLineMRTservicew erestudiedinthisresearchtoanalyzebasicdata
anddecisionc r i t e r i a todecidetousethePurpleLineMRTtowardsthedevelopmentofaBinaryCho iceModeltopredicttheprobabilityofpassengerstousethePurpleLineMRT
2.Objectives
1 TostudythedemographiccharacteristicsaffectingthedecisiontousePurpleLineMRTservice
2 TostudybehaviorsofPurpleLinepassengersaffectingpossibilityofusingPurpleLineMRTservice 3.TodevelopamathematicalmodelforpredictingtheselectionofusingPurpleLineMRTservicew henf a c t o r s andbehaviorsaffectingpassengers’decisiontousePurpleLineMRTservice
Trang 23.Methodology
Trang 3Dataanalysisinthisresearchcouldbedividedinto4partsasfollows:A n a l
y s i s ofdemographicdata
AnalysisoffactorsaffectingtheusageofPurpleLineMRTAn
al ys is ofbehaviorsofPurpleLineMRTpassengersA n a l y s
i s ofdataforhypothesistestingandsummary
- Hypothesis1:Thesampleswithdifferentdemographiccharacteristicsemphasizeonfactorsaffectingth edeci sio ntousePurpleLineMRT
- Hypothesis2:DemographiccharacteristicsofPurpleLineMRTpassengersaffectthedecisiont ouseornotusePurpleLineMRT
- Hypothesis3:FactorsaffectingthedecisiontouseornotusePurpleLineMRT
- Hypothesis4:BehaviorsofthePurpleLineMRTpassengersaffectthedecisiontouseornotusePurp leLineMRT
4 LogisticRegressionAnalysis
LogisticregressionanalysisorLogitAnalysisistheanalysisofthepredictiveequationtostudytheeff ectofapredictorvariableonadichotomousvariableorpolytomousvariable.LogisticFunctionrepr esentstherelationshipbetweenpredictorvariableandprobabilityofthe
occurrenceofeventsaccording
tothecriterionv a r i a b l e (SirichaiKanchanawasee.2005:39).Inconclusion,therearethreeconce ptsofregressionanalysisinc a s e thatdependentvariablesaredichotomousvariables(SriridejSucheew a:1996:p.17)
Conditionalmeanofregressionequationshallbeconvertedtobefrom0to1,whichissuitablefo rlogisticregressionanalysis
Thedistributionofe r r o r s mustbebinomial,w hi c h willbetheba si c s t a t i s t i c a l distributio nforf u r t h e r analysis
Otherprinciplesoflinearregressionanalysiscanbeappliedtologisticregression
.G o o d n e s s offitmustbevalidatedbytheresearcher
IfChi-squareissignificant,itmeansthata n independentvariableorasetofindependentvariablesi s re latedtothedependentvariable
If-2LogLikelihoodiscloseto0,itmeansthatthemodelhashighergoodness-of-fit thanothermodels(iIf-2LogLikelihoodiscloseto0,itmeansthatthemodelhashighergoodness-of-fitsvalueis0,itistheperfectmodel.)
ThesignificanceofeachindependentvariableinthemodelwasvalidatedwithWaldstatistic(t-statisticsw e r e usedinLinearRegression).Themodel’sabilitytoforecastisconsideredby
%ofClassification.Whenthegoodness-of-fitwa sv a l i da t e d, thenextstepsa re thea na l y s i s and determinationofstatisticsfr omthe analysis
Therelationshipbetweenasetofindependentvariablesanddependentvariablewastestedb y-2logLikelihoodandChi-squarestatistics
Themodel’sGoodnessoffitwastestedwithChi-squarestatistics
ThesignificanceofeachindependentvariableinthemodelwasvalidatedwithWaldstatistic
Themodel’spredictabilitywastestedbyconsidering%ofClassification
Studyingthepassengers’decisiontouseornotusePurpleLineMRTisusefultoprepareormanagetoplanthe implementationefficientlyandeffectively.ThevariablesaffectingthedecisiontousethePurpleLineMRTa
r e classifiedinto3groupsasfollows:
Variablesrelatedtopassengers’personalcharacteristicsaregender,ageandoccupation
Variablesrelatedtothe factorsinfluencingtheuseofthe PurpleLineMRT includingPurpleLineM R T serviceusage,fares,stationlocation,marketingpromotion,employ eeservice,serviceprocess,stationfacilities
Trang 4 VariablesrelatedtothebehaviorofusingthePurpleLineMRTincludingweeklyusagefrequency, mostfrequentlyperiodoftimetousetheservice,mainobjectiveofusingtheservice,reasontous etheservice,a n d reasonnottousetheservice
Somevariablesdescribedabovehadpositiveimpact(thisimpliedthatdatacollectedfromthequestionnair
ew e r e thenanalyzedthroughtableandLogisticRegressionAnalysis
5.Results
Theresultsofdataanalysisshowedthatthesamplesweremales(46.76%)andfemales(53.24%).Mostofthe
sampleswerethoseagedbetween21-30years(33.5%).Mostofthesampleswereemployees(33.50%).Tostudydemographicfactorsofthesample
service,11-itemquestionnairewasu s e d
TostudyfactorsaffectingthedecisiontousePurpleLineMRT,24-
itemquestionnairewasusedwith5-ratingscale.Ifanansweris“5”means“highest”leveland“1”means“lowest”level.Fromevaluatingfacto
datacollectedwerethenanalyzedtodeterminethemeanands t a n d a r d deviationasshowninTable 1
Table1:Meanandstandarddeviationofthesamplesclassifiedwithfactorsaffectingthe
decisiont o usePurpleLineMRTservice
PurpleLineMRTprovidesfastservicewithoutwaitinglong
PurpleLineMRTprovidessufficientnumberoftrainsto
PurpleLineMRTprovidesconvenienttransportofBTSorM
Passengerfaresareclassifiedbypassenger’sage.Children
andstudentsarechargedcheaperthangeneralpassengers 3.75 0.782
Alistoffaresisclearlydisplayedonticketingmachineand
Stationlocationisnearmajorlandmarkssuchasoffice,d e p
a r t m e n t storeandeducationalinstitution 3.585 0.918
Publicrelationsareavailablethroughvariousmedia 3.40 0.884
Trang 5Items x S.D Symbol
Numberofstationedofficersineachstationisenoughtom
MRTofficerspaytheirattentiontoclientsandmakethemf
StepsofusingMRTservicearenotcomplicatedandp a s s e
MRTserviceispunctual.Periodofwaitingforthepas s eng
MRTprovidef a c i l i t i e s fordisablepeoples u c h a s e n t r a n c
e e x i t , toiletforpriority,disablepeople,andBraillesigns 3.74 0.970
Elevator,toilet,telephonebox,ATM,andshopsareavailable 3.23 0.995
BothThaiandEnglishsignageareavailableforthepassengers 3.69 0.979
Tostudythe behaviorsofusingthePurple LineMRTservice,15-itemscale was
usedasshownintable2
Table2:NumberandpercentageofthesamplesclassifiedbybehaviorsofusingPurpleLineMR Tservice
Trang 6Items Frequency Percentage Symbol
Trang 709.31–12.30hrs 53 13.25
Stationisnotlocatedintheareawhereapassengerlives 190 47.50
Table3:StatisticsofthemodelofPurpleLineMRTpassengers
llR2
Nakelkerk
eR2
HosmerandLemeshowTest
Initial-2LogLikelihood:338.167
-2LLofFullModel:201.711
*Rejectthehypothesisatasignificancelevelof0.05.
FromTable3 , theC h i
-s q u a r e -stati-stic-sw a -s 1 3 6 4 5 7 (-sig.= 0 0 0 0 ) Thi-smeantthata t l e a -s t onef a c t o r influencedt
hedecisiontousePurpleLineMRTwith-2loglikelihoodvalueapproachingzero.Thisimpliedthatconstructedequationormodelhadgoodquali tyorconsistencywithdata.Cox&SnellRSquarevaluew a s 0.289whichwasnotclosetozero.Thisindicate dthemodel’sconsistencyintermsofcomparingthequalityofthemodelcreatedwiththeworstmodel,anull modelwithnoindependentvariable.TheNagelkerkeRSquarevaluewas0.507.Thismeantthatind ependentvariablescouldexplain50.70%ofthevariationinthes e r v i c e usage.WhenWaldStatist icof
over1wasconsideredandSig.valuewaslessthan0.05,onlyvariablesa s inTable4influencedpredictive equationofPurpleLineMRTusage
Trang 87
*
2.356
the
th
1.892 passengers’needs
Convenienttransportofbo
BTSorMRTlinkage
usingtheserviceand
self-service
2.955 2.818 Usagefrequencyof1-2
times/week
Usagefrequencyof3-4
time/week
during
16.633
0.834 46160.88 9
5.064 0.000
1 1
0.024*
1.000 16734186.077 LogisticRegressionEquationwasobtainedasfollows:
Table4showsthatfactorsinfluencingthedecisiontousethePurpleLineMRTwereasfollows.Sufficient numberofMRTtrainsinfluencedtheincreaseinusingPurpleLineMRTserviceby2.356times.Convenie nttransportofBTSorMRTlinkageinfluencedtheincreaseinusingPurpleLineMRTserviceby1.892ti mes.Uncomplicatedserviceusagestepsandselfserviceinfluencedtheincreaseinusing
PurpleLineMRTserviceby0.333times.QueuingofpassengersinfluencedtheincreaseinusingPurpleLineMR Tserviceby2.660times.T h e serviceusageof1-2times/
weekinfluencedtheincreaseinusingPurpleLineMRTserviceby2.995times.T h e serviceusageof3-4times/
weekinfluencedtheincreaseinusingPurpleLineMRTserviceby2.818times.T h e serviceusageduring12 31-15.30hrs.influencedtheincreaseinusingPurpleLineMRTserviceby0.153times
Trang 9usageof1-2times/weekinfluencedtheincreaseinusingPurpleLineMRTserviceby2 9 9 5 times.Thisimpliedtha
by271i n thenextfiveyearsbecauseofavailableroutes,fast,safeandconvenienttransport
Secondly,theserviceusageof3-4times/
weekinfluencedtheincreaseinusingPurpleLineMRTserviceby2.818times.Thisimpliedthatnumber ofpassengersofPurpleLineMRTwas94andwillbeincreasedby2 6 4 inthenextfiveyearsbecauseofava ilableroutes,fast,safeandconvenienttransport
Thirdly,queuingofpassengersinfluencedtheincreaseinusingPurpleLineMRTserviceby2.660ti mes.ThisimpliedthatnumberofpassengersofPurpleLineMRTwas340andwillbe increased by904inthe nextfi v e yearsbecauseofavailableroutes,fast,safeandconvenienttransport
Fourthly,sufficient numberofMRTtrainsinfluenced theincreaseinusingPurpleLineMRTservic eby
2.356times.ThisimpliedthatnumberofpassengersofPurpleLineMRTwas340andwillbeincreasedby80 1i n thenextfiveyearsbecauseofavailableroutes,fast,safeandconvenienttransport
Fifthly,convenienttransportofBTSorMRTlinkageinfluencedtheincreaseinusingPurpleLine MRTs e r v i c e by1 89 2 times.T hi s impliedt h a t numberofp a s s e n g e r s ofPu r p l e LineM RT w a
s 34 0 a n d willbei n c r e a s e d by643inthenextfiveyearsbecauseofavailableroutes,fast,safeandco nvenienttransport
Sixthly,uncomplicatedserviceusagestepsandself-serviceinfluencedtheincreaseinusingPurpleLineM R T serviceby0.333times.Thisimpliedthatnu mber
ofpassengersofPurpleLineMRTwas340andwillbeincreasedbyinthenextfiveyearsbecauseofavailabler outes,fast,safeandconvenienttransport
Seventhly,thes e r v i ce
usageduring12.31-15.30 hrs.influencedtheincreasei nusing PurpleLineMRT s e r v i c e by0.153 times Thisim pliedthatnumbero f pa ss enge rs ofPurple Li ne MRT was53 andw il l beincreasedby8innextfive years5becauseofavailableroutes,fast,safeandconvenienttransport
7.Suggestions
Morequestionnaireexamplesshouldbepreparedasgreaterinformationleadstolowererrors.Th isalsop r o v i d e s benefitstomakePurpleLineMRTserviceimprovementplan
Thestudyperiodwasfrom08.00-18.00hrs.fromMondaytoFriday.Therefore,otherperiodsoftimeshouldbeexploredtoincreasedata.Additio nalsurveyshouldbeconductedduringSaturdaysandSundaysfordatacoverage
Convenientaccesstoskytrainstationsshouldbestudied
-Greaterstudiesonpublicneedsforstationlocationsshouldbeconducted
-Morestudiesshouldbeconductedtoinvestigatewhetherthepublicacceptsfareornot
References
KamolTharuarak.
( 2005).SimulationModelofRouteSelectionBehaviorswithBinaryLogitModel(RevealedPreference).ReportofBureauofRoa dResearchandDevelopment.
KietiphongJearanuntanakij.
(2006).SimulationModelofRouteSelectionBehaviorswithMultinomialLogitModel(TheCaseStudyofBangkok-Pattaya Route).Bureauof RoadResearchandDevelopment.
ChalitaPadungmit.( 2009).FactorsAffectingtheSelectiono f
Bangkok-ChiangMaiRoutes.InternationalBusinessAdministration,Logistics&Transport, ThammasatUniversity
NapatLekawatthana.( 2013).DevelopingtheModelofSelectionofTransportforStudentsBetweenSchoolBusand OtherVehicles FacultyofTransportengineering,SuranareeUniversityofTechnology.
Trang 10TawinanSimajaruk.(2013).CostReductioninTransport:TheCaseStudyofChemical
Plant,FacultyofLogistics,BuraphaUniversity.NaruchaChaodee.
(2005).OpinionstowardstheServiceo f StateRailwayo f Thailand:T h e CaseStudyo f
theEastLine.(Bangkok-Aranyaprathet).BuraphaUniversity.
PaweenaKhampukka.(2016).TheWillingnessofStudentsandPersonneltoPayforPublicVanforUbonRatchathaniUniversity-UbonRatchathaniBus Terminal.AcademicJournal:
Far EasternUniversity.
PitiJantruthai.
(2016).FactorsInfluencingtheDecisiontoSelectPublicSystemintheCityArea.AcademicJournal:PrincessofNaradhiwasUnive rsity.
Trang 11PacharaJitchang.(2010).TheStudyofDecisiontoUseMRTofPeopleInBangkok.General Business Management
SilpakornUniversity.B h a w a t Chaichannawatik.
(2013).LongDistanceTravelBehaviorso f BangkokResidents:T h e Studyo f SelectingTransportModel.
AcademicJournal.KasemBunditUniversity)
YuthKaiyawan.
( 2012).Principlesa n d Applicationso f LogisticRegressionAnalysisforResearch.ResearchJournal:RajamangalaUniversityo fTechnologySrivijaya.
WeerayuthWattanatham.
(2012).FactorsInfluencingtheDecisionofPassengerstoSelecttheRailwayTransportServiceofSuvarnabhumiAirportandthe AirportTerminal intheCity.DepartmentofCivil Engineering.FacultyofEngineering.KasetsartUniversity.
Sirichai Karnchanawasee.(2005).Multi-levelAnalysis.Bangkok:3 rd Edition.ChulalongkornUniversityPress.
SiridejSujiva.(1996,January-June).LogisticsLogisticAnalysis:Concept,
Analysis,andInterpretationResearchMethodology.8(1):10-34.
AtcharaphornJirachartphong.
(2013).UsingRailwayTransportServiceofSuvarnabhumiAirportandtheAirportTerminalintheCityinPassengers’Perspective Departmentof
MarketingManagement.Facultyof Business Administration.DhurakijPunditUniversity.