Mode-Wise Corridor Level Travel-Time Estimation Using Machine Learning Models

Document Type : Regular Article


1 Research Scholar, Transportation Systems Engineering, Civil Engineering Department, Indian Institute of Technology Bombay, Mumbai, India

2 Associate Professor, Transportation Systems Engineering, Civil Engineering Department, Indian Institute of Technology Bombay, Mumbai, India


This research is oriented towards exploring mode-wise corridor level travel-time estimation using Machine learning techniques such as Artificial Neural Network (ANN) and Support Vector Machine (SVM). Authors have considered buses (equipped with in-vehicle GPS) as the probe vehicles and attempted to calculate the travel-time of other modes such as cars along a stretch of arterial roads. The proposed study considers various influential factors that affect travel time such as road geometry, traffic parameters, location information from the GPS receiver and other spatiotemporal parameters that affect the travel-time. The study used a segment modeling method for segregating the data based on identified bus stop locations. A k-fold cross-validation technique was used for determining the optimum model parameters to be used in the ANN and SVM models. The developed models were tested on a study corridor of 59.48 km stretch in Mumbai, India. The data for this study were collected for a period of five days (Monday-Friday) during the morning peak period (from 8.00 am to 11.00 am). Evaluation scores such as MAPE (mean absolute percentage error), MAD (mean absolute deviation) and RMSE (root mean square error) were used for testing the performance of the models. The MAPE values for ANN and SVM models are 11.65 and 10.78 respectively. The developed model is further statistically validated using the Kolmogorov-Smirnov test. The results obtained from these tests proved that the proposed model is statistically valid.


Google Scholar


Main Subjects

[1]     Konbattulwar V, Velaga NR, Jain S, Sharmila RB. Development of in-vehicle noise prediction models for Mumbai Metropolitan Region, India. J Traffic Transp Eng (English Ed 2016;3:380–7. doi:10.1016/j.jtte.2016.04.002.
[2]     Hess DB. Access to Employment for Adults in Poverty in the Buffalo-Niagara Region. Urban Stud 2005;42:1177–200. doi:10.1080/00420980500121384.
[3]     Kawabata M. Job Access and Employment among Low-Skilled Autoless Workers in US Metropolitan Areas. Environ Plan A Econ Sp 2003;35:1651–68. doi:10.1068/a35209.
[4]     Kawabata M, Shen Q. Commuting Inequality between Cars and Public Transit: The Case of the San Francisco Bay Area, 1990-2000. Urban Stud 2007;44:1759–80. doi:10.1080/00420980701426616.
[5]     Levinson DM. Accessibility and the journey to work. J Transp Geogr 1998;6:11–21. doi:10.1016/S0966-6923(97)00036-7.
[6]     Silva C, Pinho P. The Structural Accessibility Layer (SAL): Revealing how Urban Structure Constrains Travel Choice. Environ Plan A Econ Sp 2010;42:2735–52. doi:10.1068/a42477.
[7]     Kwok RCW, Yeh AGO. The Use of Modal Accessibility Gap as an Indicator for Sustainable Transport Development. Environ Plan A Econ Sp 2004;36:921–36. doi:10.1068/a3673.
[8]     Salonen M, Toivonen T. Modelling travel time in urban networks: comparable measures for private car and public transport. J Transp Geogr 2013;31:143–53. doi:10.1016/j.jtrangeo.2013.06.011.
[9]     Lei TL, Church RL. Mapping transitā€based access: integrating GIS, routes and schedules. Int J Geogr Inf Sci 2010;24:283–304. doi:10.1080/13658810902835404.
[10]    Liu S, Zhu X. Accessibility Analyst: An Integrated GIS Tool for Accessibility Analysis in Urban Transportation Planning. Environ Plan B Plan Des 2004;31:105–24. doi:10.1068/b305.
[11]    Moniruzzaman M, Páez A. Accessibility to transit, by transit, and mode share: application of a logistic model with spatial filters. J Transp Geogr 2012;24:198–205. doi:10.1016/j.jtrangeo.2012.02.006.
[12]    O’Sullivan D, Morrison A, Shearer J. Using desktop GIS for the investigation of accessibility by public transport: an isochrone approach. Int J Geogr Inf Sci 2000;14:85–104. doi:10.1080/136588100240976.
[13]    Mavoa S, Witten K, McCreanor T, O’Sullivan D. GIS based destination accessibility via public transit and walking in Auckland, New Zealand. J Transp Geogr 2012;20:15–22. doi:10.1016/j.jtrangeo.2011.10.001.
[14]    Jeong R, Rilett R. Bus arrival time prediction using artificial neural network model. Proceedings. 7th Int. IEEE Conf. Intell. Transp. Syst. (IEEE Cat. No. 04TH8749), IEEE; 2004, p. 988–93.
[15]    Ramakrishna Y, Ramakrishna P, Lakshmanan V, Sivanandan R. Bus travel time prediction using GPS data. Proc Map India 2006.
[16]    Chien SI-J, Kuchipudi CM. Dynamic Travel Time Prediction with Real-Time and Historic Data. J Transp Eng 2003;129:608–16. doi:10.1061/(ASCE)0733-947X(2003)129:6(608).
[17]    Fan W, Gurmu Z. Dynamic Travel Time Prediction Models for Buses Using Only GPS Data. Int J Transp Sci Technol 2015;4:353–66. doi:10.1016/S2046-0430(16)30168-X.
[18]    Rajbhandari R. Bus arrival time prediction using stochastic time series and Markov chains 2004.
[19]    Suwardo W, Napiah M, Kamaruddin I. ARIMA models for bus travel time prediction. J Inst Eng Malaysia 2010:49–58.
[20]    Rahmani M, Jenelius E, Koutsopoulos HN. Route travel time estimation using low-frequency floating car data. 16th Int. IEEE Conf. Intell. Transp. Syst. (ITSC 2013), IEEE; 2013, p. 2292–7. doi:10.1109/ITSC.2013.6728569.
[21]    Jenelius E, Koutsopoulos HN. Travel time estimation for urban road networks using low frequency probe vehicle data. Transp Res Part B Methodol 2013;53:64–81. doi:10.1016/j.trb.2013.03.008.
[22]    Cathey FW, Dailey DJ. A prescription for transit arrival/departure prediction using automatic vehicle location data. Transp Res Part C Emerg Technol 2003;11:241–64. doi:10.1016/S0968-090X(03)00023-8.
[23]    Wall Z, Dailey DJ. An algorithm for predicting the arrival time of mass transit vehicles using automatic vehicle location data. 78th Annu. Meet. Transp. Res. board, Citeseer; 1999, p. 1–11.
[24]    Shalaby A, Farhan A. Prediction Model of Bus Arrival and Departure Times Using AVL and APC Data. J Public Transp 2004;7:41–61. doi:10.5038/2375-0901.7.1.3.
[25]    Chen H, Rakha HA. Real-time travel time prediction using particle filtering with a non-explicit state-transition model. Transp Res Part C Emerg Technol 2014;43:112–26. doi:10.1016/j.trc.2014.02.008.
[26]    Dhivyabharathi B, Anil Kumar B, Vanajakshi L, Panda M. Particle Filter for Reliable Bus Travel Time Prediction Under Indian Traffic Conditions. Transp Dev Econ 2017;3:13. doi:10.1007/s40890-017-0043-z.
[27]    Bi J, Chang C, Fan Y. Particle Filter for Estimating Freeway Traffic State in Beijing. Math Probl Eng 2013;2013:1–6. doi:10.1155/2013/382042.
[28]    Vanajakshi L, Rilett LR. Support Vector Machine Technique for the Short Term Prediction of Travel Time. 2007 IEEE Intell. Veh. Symp., IEEE; 2007, p. 600–5. doi:10.1109/IVS.2007.4290181.
[29]    Hofleitner A, Herring R, Bayen A. Arterial travel time forecast with streaming data: A hybrid approach of flow modeling and machine learning. Transp Res Part B Methodol 2012;46:1097–122. doi:10.1016/j.trb.2012.03.006.
[30]    Zheng F, Van Zuylen H. Urban link travel time estimation based on sparse probe vehicle data. Transp Res Part C Emerg Technol 2013;31:145–57. doi:10.1016/j.trc.2012.04.007.
[31]    Li L, He S, Zhang J, Ran B. Short-term highway traffic flow prediction based on a hybrid strategy considering temporal-spatial information. J Adv Transp 2016;50:2029–40. doi:10.1002/atr.1443.
[32]    Wang L, Zuo Z, Fu J. Bus arrival time prediction using RBF neural networks adjusted by online data. Procedia-Social Behav Sci 2014;138:67–75.
[33]    Zhang Z, Wang Y, Chen P, He Z, Yu G. Probe data-driven travel time forecasting for urban expressways by matching similar spatiotemporal traffic patterns. Transp Res Part C Emerg Technol 2017;85:476–93. doi:10.1016/j.trc.2017.10.010.
[34]    Yu B, Lam WHK, Tam ML. Bus arrival time prediction at bus stop with multiple routes. Transp Res Part C Emerg Technol 2011;19:1157–70. doi:10.1016/j.trc.2011.01.003.
[35]    Akter S, Huda T, Nahar L, Akter S. Travel time prediction using support vector machine (SVM) and weighted moving average (WMA). Int J Eng Res Technol 2015;4:496–503.
[36]    Chakroborty P, Kikuchi S. Using Bus Travel Time Data to Estimate Travel Times on Urban Corridors. Transp Res Rec J Transp Res Board 2004;1870:18–25. doi:10.3141/1870-03.
[37]    Patnaik J, Chien S, Bladikas A. Estimation of bus arrival times using APC data. J Public Transp 2004;7:1.
[38]    Chang H, Park D, Lee S, Lee H, Baek S. Dynamic multi-interval bus travel time prediction using bus transit data. Transportmetrica 2010;6:19–38. doi:10.1080/18128600902929591.
[39]    Zhou Y, Yao L, Chen Y, Gong Y, Lai J. Bus arrival time calculation model based on smart card data. Transp Res Part C Emerg Technol 2017;74:81–96. doi:10.1016/j.trc.2016.11.014.
[40]    Al-Deek HM, D’Angelo MP, Wang MC. Travel time prediction with non-linear time series. Fifth Int. Conf. Appl. Adv. Technol. Transp. Eng., 1998.
[41]    Stathopoulos A, Karlaftis MG. A multivariate state space approach for urban traffic flow modeling and prediction. Transp Res Part C Emerg Technol 2003;11:121–35. doi:10.1016/S0968-090X(03)00004-4.
[42]    Chien SI-J, Ding Y, Wei C. Dynamic Bus Arrival Time Prediction with Artificial Neural Networks. J Transp Eng 2002;128:429–38. doi:10.1061/(ASCE)0733-947X(2002)128:5(429).
[43]    Xia J, Chen M, Huang W. A multistep corridor travel-time prediction method using presence-type vehicle detector data. J Intell Transp Syst Technol Planning, Oper 2011;15:104–13. doi:10.1080/15472450.2011.570114.
[44]    Chen P, Ding C, Lu G, Wang Y. Short-term traffic states forecasting considering spatial-temporal impact on an urban expressway. Transp Res Rec 2016;2594. doi:10.3141/2594-10.
[45]    Park D, Rilett LR. Forecasting Freeway Link Travel Times with a Multilayer Feedforward Neural Network. Comput Civ Infrastruct Eng 1999;14:357–67. doi:10.1111/0885-9507.00154.
[46]    Kumar BA, Vanajakshi L, Subramanian SC. Bus travel time prediction using a time-space discretization approach. Transp Res Part C Emerg Technol 2017;79:308–32. doi:10.1016/j.trc.2017.04.002.
[47]    Cherrett T, Mcleod F, Bell H, McDonald M. Journey time estimation using single inductive loop detectors on non-signalised links. J Oper Res Soc 2002;53:610–9. doi:10.1057/palgrave.jors.2601348.
[48]    Jeong R, Rilett R. Bus arrival time prediction using artificial neural network model. Proceedings. 7th Int. IEEE Conf. Intell. Transp. Syst. (IEEE Cat. No.04TH8749), IEEE; n.d., p. 988–93. doi:10.1109/ITSC.2004.1399041.
[49]    Dharia A, Adeli H. Neural network model for rapid forecasting of freeway link travel time. Eng Appl Artif Intell 2003;16:607–13. doi:10.1016/j.engappai.2003.09.011.
[50]    Yin T, Zhong G, Zhang J, He S, Ran B. A prediction model of bus arrival time at stops with multi-routes. Transp Res Procedia 2017;25:4623–36. doi:10.1016/j.trpro.2017.05.381.
[51]    Vanajakshi L, Rilett LR. A comparison of the performance of artificial. neural networks and support vector machines for the prediction of traffic speed. IEEE Intell. Veh. Symp. 2004, IEEE; n.d., p. 194–9. doi:10.1109/IVS.2004.1336380.
[52]    Zhong S., Hu J b., Ke S., Wang X., Zhao J., Yao B. A Hybrid Model based on Support Vector Machine for Bus Travel-Time Prediction. Promet - Traffic - Traffico 2015;27:291–300. doi:10.7307/ptt.v27i4.1577.
[53]    Muller K-R, Mika S, Ratsch G, Tsuda K, Scholkopf B. An introduction to kernel-based learning algorithms. IEEE Trans Neural Networks 2001;12:181–201. doi:10.1109/72.914517.
[54]    Bin Y, Zhongzhen Y, Baozhen Y. Bus Arrival Time Prediction Using Support Vector Machines. J Intell Transp Syst 2006;10:151–8. doi:10.1080/15472450600981009.
[55]    Liu H, Van Zuylen H, Van Lint H, Salomons M. Predicting Urban Arterial Travel Time with State-Space Neural Networks and Kalman Filters. Transp Res Rec J Transp Res Board 2006;1968:99–108. doi:10.1177/0361198106196800112.
[56]    van Lint JWC, Hoogendoorn SP, van Zuylen HJ. Accurate freeway travel time prediction with state-space neural networks under missing data. Transp Res Part C Emerg Technol 2005;13:347–69. doi:10.1016/j.trc.2005.03.001.
[57]    Kisgyörgy L, Rilett LR. Travel time prediction by advanced neural network. Period Polytech Civ Eng 2002;46:15–32.
[58]    Ishak S, Alecsandru C. Optimizing Traffic Prediction Performance of Neural Networks under Various Topological, Input, and Traffic Condition Settings. J Transp Eng 2004;130:452–65. doi:10.1061/(ASCE)0733-947X(2004)130:4(452).
[59]    Sanghoon Bae, Kachroo P. Proactive travel time predictions under interrupted flow condition. Pacific Rim TransTech Conf. 1995 Veh. Navig. Inf. Syst. Conf. Proceedings. 6th Int. VNIS. A Ride into Futur., IEEE; n.d., p. 179–86. doi:10.1109/VNIS.1995.518836.
[60]    Hall RW, Vyas N. Buses as a Traffic Probe: Demonstration Project. Transp Res Rec J Transp Res Board 2000;1731:96–103. doi:10.3141/1731-12.
[61]    Tantiyanugulchai S, Bertini RL. Arterial performance measurement using transit buses as probe vehicles. Proc. 2003 IEEE Int. Conf. Intell. Transp. Syst., IEEE; n.d., p. 102–7. doi:10.1109/ITSC.2003.1251929.
[62]    Bertini RL, Tantiyanugulchai S. Transit Buses as Traffic Probes: Use of Geolocation Data for Empirical Evaluation. Transp Res Rec J Transp Res Board 2004;1870:35–45. doi:10.3141/1870-05.
[63]    Padmanaban RPS, Vanajakshi L, Subramanian SC. Automated Delay Identification for Bus Travel Time Prediction towards APTS Applications. 2009 Second Int. Conf. Emerg. Trends Eng. Technol., IEEE; 2009, p. 564–9. doi:10.1109/ICETET.2009.43.
[64]    Vasantha Kumar S, Vanajakshi L. Urban Arterial Travel Time Estimation Using Buses as Probes. Arab J Sci Eng 2014;39:7555–67. doi:10.1007/s13369-014-1332-z.
[65]    Shalaby A, Farhan A. Bus Travel Time Prediction Model for Dynamic Operations Control and Passenger Information Systems. Transp. Res. Board 82nd Annu. Meet., 2003.
[66]    El Esawey M, Sayed T. A framework for neighbour links travel time estimation in an urban network. Transp Plan Technol 2012;35:281–301. doi:10.1080/03081060.2012.671028.
[67]    Bae S. Dynamic estimation of travel time on arterial roads by using automatic vehicle location (AVL) bus as a vehicle probe 1995.
[68]    Chakroborty P, Kikuchi S. Using bus travel time data to estimate travel times on urban corridors. Transp Res Rec J Transp Res Board 2004:18–25.
[69]    Jeong R, Rilett LR. Prediction Model of Bus Arrival Time for Real-Time Applications. Transp Res Rec J Transp Res Board 2005;1927:195–204. doi:10.1177/0361198105192700123.
[70]    Padmanaban RPS, Divakar K, Vanajakshi L, Subramanian SC. Development of a real-time bus arrival prediction system for Indian traffic conditions. IET Intell Transp Syst 2010;4:189. doi:10.1049/iet-its.2009.0079.
[71]    Kieu LM, Bhaskar A, Chung E. Bus and car travel time on urban networks: integrating bluetooth and bus vehicle identification data 2012.
[72]    Zhan X, Hasan S, Ukkusuri S V., Kamga C. Urban link travel time estimation using large-scale taxi data with partial information. Transp Res Part C Emerg Technol 2013;33:37–49. doi:10.1016/j.trc.2013.04.001.
[73]    Arhin S, Stinson RZ. Transit Bus Travel Time Prediction using AVL Data. Int J Eng Res Technol 2016;5:21–6.
[74]    Sharmila RB, Velaga NR, Kumar A. SVM-based hybrid approach for corridor-level travel-time estimation. IET Intell Transp Syst 2019;13:1429–39. doi:10.1049/iet-its.2018.5069.
[75]    Mattera D, Haykin S. Support vector machines for dynamic reconstruction of a chaotic system. Adv. kernel methods Support vector Learn., 1999, p. 211–41.
[76]    Reddy KK, Kumar BA, Vanajakshi L. Bus travel time prediction under high variability conditions. Curr Sci 2016:700–11.
[77]    Pawar DS, Patil GR, Chandrasekharan A, Upadhyaya S. Classification of Gaps at Uncontrolled Intersections and Midblock Crossings Using Support Vector Machines. Transp Res Rec J Transp Res Board 2015;2515:26–33. doi:10.3141/2515-04.
[78]    Vapnik VN. Computational Learning Theory. 1998.
[79]    Burges CJC. A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov 1998;2:121–67.
[80]    Smola AJ, Schölkopf B. A tutorial on support vector regression. Stat Comput 2004;14:199–222. doi:10.1023/B:STCO.0000035301.49549.88.
[81]    Cortes C, Vapnik V. Support-vector networks. Mach Learn 1995;20:273–97. doi:10.1007/BF00994018.
[82]    Nagalla R, Pothuganti P, Pawar DS. Analyzing Gap Acceptance Behavior at Unsignalized Intersections Using Support Vector Machines, Decision Tree and Random Forests. ANT/SEIT, 2017, p. 474–81.
[83]    Kenneth DL, Ronald KK. Advances in business and management forecasting. Bingley, UK Emerald Books 1982.