Abstract
The use of managed lanes (MLs) is currently predicted using discrete choice models. However, recent research discovered that many travelers were not making choices in their use of MLs—they always used one set of lanes and never varied. Therefore, choice models may not be appropriate. This research examined if machine learning could be used to model ML use by travelers. Machine learning techniques were used to classify approximately 125,000 travelers who made approximately 3.5 million trips on the Katy Freeway over a 2-year period. Different machine learning models were able to classify travelers into those who always used MLs, always used toll-free general purpose lanes (GPLs), and those who used both types of lanes (choosers). Travel time savings, total number of trips made, and the start and end location of the trip were key predictors of the user class. Neural networks with LSTM could classify the trips conducted by choosers as ML trips or GPL trips based on the start and end location of the trips and the drivers’ travel history. A comparison with logit models showed that machine learning models performed better in classifying travelers and trips made by choosers. However, trends obtained from both models were same. This research has shown that a multi-level classification of travelers and machine learning techniques were able to predict real-world traveler choice on a ML facility. This may be an important first step in shifting travel models from how we think travelers are behaving to what is actually happening.
Similar content being viewed by others
References
Abir AKM, Burris M, Spiegelman C (2017) The value of travel time and reliability: empirical evidence from Katy freeway. Transp Res Rec 2606:71–78
Alemazkoor N, Burris M, Danda S (2015) Using empirical data to find the best measure of travel time reliability. J Transp Res Board. https://doi.org/10.3141/2530-11,pp.93-100
Breiman L (2001) Random forests. Mach Learn 45(1):5–32. https://doi.org/10.1023/a:1010933404324
Brownlee J (2018) A gentle introduction to k-fold cross-validation. Machine Learning Mastery. https://machinelearningmastery.com/k-fold-cross-validation/. Accessed Nov 2020
Burris M, Brady J (2018) Unrevealed preferences: unexpected traveler response to pricing on managed lanes. J Transp Res Board. https://doi.org/10.1177/0361198118796928
Chang CC, Lin CJ (2011) LIBSVM. ACM Trans Intell Syst Technol 2(3):1–27. https://doi.org/10.1145/1961189.1961199
Chollet F (2017) Deep learning with Python. Simon and Schuster
FHWA (2004) Managed lanes: a cross-cutting study. https://ops.fhwa.dot.gov/freewaymgmt/publications/managed_lanes/crosscuttingstudy/final3_05.pdf. Accessed Jun 2020
Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 1189–1232
Goldstein A et al (2015) Peeking inside the black box: visualizing statistical learning with plots of individual conditional expectation. J Comput Graph Stat 24(1):44–65
Hagenauer J, Helbich M (2017) A comparative study of machine learning classifiers for modeling travel mode choice. Expert Syst Appl 78:273–282
Hess S, Daly A, Batley R (2018) Revisiting consistency with random utility maximisation: theory and implications for practical work. Theor Decis 84(2):181–204
Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780
Hsu CW, Lin CJ (2002) A comparison of methods for multiclass support vector machines. IEEE Trans Neural Netw 13(2):415–425
Karlsen MR, Moschoyiannis S (2018) Learning condition–action rules for personalised journey recommendations. In: International joint conference on rules and reasoning. Springer, Cham, pp 293–301
Ketkar N (2017) Introduction to keras. In: Deep learning with Python. Apress Berkeley, CA. https://doi.org/10.1007/978-1-4842-2766-4_7
Koller D, Sahami M (1996) Toward optimal feature selection. Stanford InfoLab. http://ilpubs.stanford.edu:8090/208/1/1996-77.pdf
Koppelman FS, Bhat C (2006) A self-instructing course in mode choice modeling: multinomial and nested logit models. https://www.caee.utexas.edu/prof/bhat/COURSES/LM_Draft_060131Final-060630.pdf. Accessed May 2019
Menze BH, Kelm BM, Masuch R, Himmelreich U, Bachert P, Petrich W, Hamprecht FA (2009) A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data. BMC Bioinform 10(1):213. https://doi.org/10.1186/1471-2105-10-213
Nakayama S, Kitamura R, Fujii S (2001) Drivers’ route choice rules and network behavior: do drivers become rational and homogeneous through learning? Transp Res Rec 1752(1):62–68
Nam D, Kim H, Cho J, Jayakrishnan R (2017) A model based on deep learning for predicting travel mode choice. In: Proceedings of the transportation research board 96th annual meeting transportation research board, Washington, DC, pp 8–12
Omrani H (2015) Predicting travel mode of individuals by machine learning. Transp Res Proc 10:840–849
Patil S, Burris M, Shaw D (2011) Travel using managed lanes: an application of a stated choice model for Houston, Texas. Transp Policy 18(2011):595–603
Pirra M, Diana M (2019) A study of tour-based mode choice based on a support vector machine classifier. Transp Planning Technol 42(1):23–36
Sasaki M (2007) The truth of the F-measure. Teach Tutor Mater. https://www.cs.odu.edu/~mukka/cs795sum09dm/Lecturenotes/Day3/F-measure-YS-26Oct07.pdf
Sharifi F, Burris MW (2019) Application of machine learning to characterize uneconomical managed lane choice behaviour. Case Stud Transp Policy 7(4):781–789
Simon HA (1990) Invariants of human behavior. Annu Rev Psychol 41(1):1–20
Sun L, Karwan MH, Kwon C (2016) Incorporating driver behaviors in network design problems: challenges and opportunities. Transp Rev 36(4):454–478
Suriya Prakash J, Annamalai Vignesh K, Ashok C, Adithyan R (2012) Multi class support vector machines classifier for machine vision application. In: Paper presented at the 2012 international conference on machine vision and image processing (MVIP)
Towards Data Science (2019) A look into feature importance in logistic regression models. https://towardsdatascience.com/a-lookinto-feature-importance-in-logistic-regression-models-a4aa970f9b0f. Accessed Dec 2019
Xie C, Lu J, Parkany E (2003) Work travel mode choice modeling with data mining: decision trees and neural networks. Transp Res Rec 1854(1):50–61
Xumei C, Huibo G, Wang J (2012) BRT vehicle travel time prediction based on SVM and Kalman filter. J Transp Syst Eng Inf Technol 12(4):29–34
Zhang Y, Xie Y (2008) Travel mode choice modeling with support vector machines. Transp Res Rec 2076(1):141–150
Zhao X, Yan X, Van Hentenryck P (2019) Modeling heterogeneity in mode-switching behavior under a mobility-on-demand transit system: an interpretable machine learning approach. arXiv:1902.02904
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ashraf, S., Abianeh, A.S., Sharifi, F. et al. Predicting the Use of Managed Lanes Using Machine Learning. J. Big Data Anal. Transp. 3, 213–227 (2021). https://doi.org/10.1007/s42421-021-00048-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42421-021-00048-x