曹祎 朱海
摘要:手機召車軟件深刻變革了城市出租車系統(tǒng),為供需雙方搭建了信息平臺,出租車空駛時,需完成雙層選擇。上層為選擇使用手機召車軟件的網約模式或者不使用手機召車軟件的巡游模式;下層為選擇下一乘客的起點小區(qū),下層區(qū)域選擇受上層模式選擇特征影響。文章分析了模式選擇與區(qū)域選擇的關系,構建基于出租車選擇行為的出租車運營平衡模型,并探討兩種模式對應的乘客在等待過程上的差異,以便于計算基于小區(qū)特征的等待時間。算例結果表明:網約模式的出租車搜索時間比巡游模式受出租車規(guī)模影響小;出租車規(guī)模較小時,巡游模式的利用率較網約模式低;隨著規(guī)模增大,巡游模式利用率下降6.2%,網約模式利用率下降13.1%,且此時網約模式的利用率較低,乘客等待時間的差距縮短??梢?,手機召車軟件可有效改善信息不對稱,但在出租車規(guī)模較大時邊際效應明顯。
關鍵詞:城市出租車系統(tǒng);均衡模型;手機召車軟件;出租車搜索行為;乘客等待時間
The ehailing software has profoundly changed the urban taxi system,and builds an information platform for both the supply and demand sides,and,when the taxi is empty,it needs to complete the doublelayer selection,the upper layer is to select the online reservation mode of ehailing software or not using the cruise mode of ehailing software,the lower layer is to select the starting community of next passenger,and the lowerlayer area selection is affected by the upperlayer mode selection feature.This article analyzes the relationship between mode selection and region selection,constructs a taxi operation equilibrium model based on taxi selection behavior,and discusses the difference of corresponding passenger waiting process under the two modes,in order to calculate the waiting time based on community characteristics.The calculation results show that the taxi search time of online reservation mode is less affected by the taxi scale than the cruise mode.When the taxi scale is small,the utilization rate of cruise mode is lower than online reservation mode.As the scale increases,the cruise mode utilization rate is decreased by 6.2%,and the utilization rate of online reservation mode is decreased by 13.1%.At this time,the utilization rate of online reservation mode is lower,and the gap between passenger waiting times is shortened.It can be seen that the ehailing software can effectively improve the information asymmetry,but the marginal effect is obvious when the taxi scale is large
Urban taxi system;Equilibrium model;Ehailing software;Taxi search behavior;Passenger waiting time
0 引言
“互聯網+”城市交通為城市公共出行服務的供給側改革和優(yōu)化帶來了機遇,需要重新審視基于信息平臺的城市出租車系統(tǒng)。面對技術革新與行業(yè)演化,急需理清出租車系統(tǒng)中傳統(tǒng)與新型兩種模式的差異,把握信息背景下的市場運轉機制特征。
Yang和Wong在Douglas[1]提出在經典出租車固定需求的框架上開創(chuàng)出租車市場均衡問題[2-3]。Wong
認為空駛出租車的選擇由兩個子模型構成[4-5]。以上研究均設定在傳統(tǒng)出租車市場的背景下以討論空駛行為的隨機性。然而在信息環(huán)境下,空駛車可通過手機召車軟件提前獲知需求分布信息,基于自身利益最大化動態(tài)調整搜索策略,傳統(tǒng)研究難以完整涵括這種轉變過程。
目前,已有學者注意到出租車市場變化態(tài)勢:Zhao采用紐約的出租車信息平臺數據,得出手機召車軟件這一技術創(chuàng)新為制度革新提供了可行性[6];Xia構建了出租車行業(yè)評價指標體系[7];Hai認為互聯網專車是出租車的有效補充[8];He討論了出租車司機及乘客的雙邊選擇均衡問題[9]。以上研究多集中于對手機召車軟件這一技術手段本身的探討,較少涉及系統(tǒng)運轉機制及內生變量的變化。
通過以上分析可知,空駛出租車的搜索行為,是區(qū)分新型與傳統(tǒng)出租車系統(tǒng)的關鍵。將使用手機召車軟件完成出租車服務的召車模式稱為網約模式,反之為巡游模式。假設所有出租車均已配備手機召車軟件,在上一乘客下車后,可自由選擇下一乘客。本文擬量化出租車選擇行為,構建信息環(huán)境下出租車系統(tǒng)重構的運營平衡模型,旨在為決策部門掌握手機召車軟件特性及制定管控方案提供理論參考。
隨著出租車規(guī)模增加,結合圖2(a)、圖2(c)可知:巡游車搜索時間總增幅為51%,出租車規(guī)模增加意味著市場中出租車運力增加,即總運營時間增加,在需求固定的情況下,巡游車搜索對象為路網中的任一乘客,因此搜索時間增加,整體利用率降低了6.2%;網約車搜索時間不受出租車規(guī)模影響,這是由于網約車的服務對象為路網中的特定的已預約乘客,搜索時間只由供需雙方的距離等路網特征決定,搜索時間穩(wěn)定,整體利用率降低了13.1%,比巡游車更為敏感。由圖2(b)可知:隨著出租車規(guī)模的增加,巡游車乘客更容易揚招到出租車,等待時間下降明顯;參考式(7)、式(8)可知,隨著巡游車的負效用增加,而網約車搜索時間未變,網約車的負效用也不變,出租車更傾向于選擇網約模式;由式(18)可知,網約車乘客等待時間則間接降低。此外,在出租車規(guī)模較小時,網約模式的乘客等待時間短,同時利用率也高;規(guī)模較大時,網約車與巡游車乘客等待時間接近,此時巡游模式的利用率較高,說明此時手機召車軟件邊際效應有限。
5 結語
在手機召車軟件廣泛使用的背景下,本文確定了空駛出租車選擇行為的多層次結構。上層為模式選擇(巡游模式或網約模式);下層為區(qū)域選擇(下一乘客的起點小區(qū))。并依次量化了選擇效用,建立了考慮空間路網特征的出租車市場平衡模型。在平衡狀態(tài)時,空駛出租車不可通過改變選擇行為以增加效用。通過算例分析可知,在一定的規(guī)模下,巡游模式在出租車搜索時間及乘客等待時間上,受租車規(guī)模影響的波動性更大,證實了手機召車軟件在規(guī)模較小時對信息不對稱的改善作用;但在規(guī)模較大時,該作用十分有限。本文的研究是在出租車總需求固定的前提下進行的,對于彈性需求的變化,將是下一步的研究方向。
參考文獻:
[1]Douglas G W,Douglas G W.Price Regulation and Optimal Service Standards:The taxicab Industry [J].Journal of Transport Economics and Policy,1972(20):116-127.
[2]Yang H,Wong S C.A network model of urban taxi services [J].Transportation Research Part B Methodological,1998,32(4):235-246.
[3]Yang H,Wong K I,Wong S C.Modeling urban taxi services in road networks:Progress,problem and prospect [J].Journal of Advanced Transportation,2001,35(3):237-258.
[4]Wong R C P,Szeto W Y,Wong S C.A twostage approach to modeling vacant taxi movements [J].Transportation Research Part C Emerging Technologies,2015(59):147-163.
[5]Wong R C P,Szeto W Y,Wong S C.Behavior of taxi customers in hailing vacant taxis:a nested logit model for policy analysis:[J].Journal of Advanced Transportation,2015,49(8):867-883.
[6]Zhao J.Humanizing Travel:How Ehail Apps Transform Stakeholder Relationships in the U.S.Taxi Industry[C].Transportation Research Board 94th Annual Meeting,2015.
[7]Xia T.Evaluation of Efficiency of Taxi Industry:Effects of TaxiCalling Mobile Apps[C].Transportation Research Board Annual Meeting 94th Annual Meeting,2015.
[8]Hai D J,Yang Y J,Li W F,et al.How the Chauffeured Cars Affect the Taxi Market? A Case Study in Xiamen,China[C].Presented at the 95th Annual Meeting of the Transportation Research Board,Washington,D.C.,2016.
[9]He F,Shen Z J M.Modeling taxi services with smartphonebased ehailing applications [J].Transportation Research Part C Emerging Technologies,2015(58):93-106.
[10]Wang X L,He F,Gao O.Pricing Strategies for an Ehailing Platform in Taxi Service [C].Presented at the 95th Annual Meeting of the Transportation Research Board,Washington,D.C.,2016.
[11]曹 祎,陶竑宇,羅 霞.打車軟件使用率對出租車社會福利的影響[J].交通運輸系統(tǒng)工程與信息,2015,15(3):1-6.
[12]Shi Y,Lian Z.Optimization and Strategic Behavior in a Passengertaxi Service System[J].European Journal of Operational Research,2016(249):1024-1032.