Predicting the Use of Managed Lanes Using Machine Learning

Ashraf, Sruthi; Abianeh, Arezoo Samimi; Sharifi, Farinoush; Gupta, Vivek; Narsaria, Isha Shyam; Burris, Mark

doi:10.1007/s42421-021-00048-x

Cited by 3 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Not all travelers are ready to do the travel time versus toll analysis as the existing pricing models assume. Previous research has shown the need for classification of ML users ( 1 , 54 ). This research is an initial step toward fulfilling that need.…”

Section: Discussionmentioning

confidence: 99%

Using Behavioral Economics to Identify Potential Managed Lane Users

Ashraf

Burris

Brown

et al. 2022

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

Priced managed lanes (MLs) offer travelers an option to choose to pay a toll to travel on the MLs for a generally faster, more reliable travel than on the adjacent, toll-free general-purpose lanes (GPLs). Recent research has shown that many travelers on freeways with MLs choose the same lane type every trip regardless of travel time savings and toll rate. In this paper, a classic experimental economics traffic game was replicated and re-examined to model traveler choice by classifying travelers as “choosers” (people who frequently choose between MLs and GPLs) and “non-choosers” (people who use only one type of lane). This traffic experiment was then augmented with a travel survey to find the individual differences (psychological traits), trip-related variables, and socio-demographic variables that help differentiate travelers into the above two categories. Based on the traffic experiment and travel survey, it was found that: (1) the experiment could identify real life choosers and (2) many more travelers indicate that they are choosers in a survey than in actual travel. Features that are related to the choosing behavior were the time taken by the subject to answer a survey section and time used to verify their answers before submitting. Participants who spent more time on those things tended to be choosers. Travelers who were direct responders in the experiment tended to be choosers. Other factors that differentiated choosers from non-choosers were trip duration, familiarity with the ML facilities, and education level.

show abstract

Section: Discussionmentioning

confidence: 99%

Using Behavioral Economics to Identify Potential Managed Lane Users

Ashraf

Burris

Brown

et al. 2022

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

show abstract

“…A dynamic lane must be managed in order to enhance traffic fluidity, to ensure services such as: the increase and optimization of roadway capacity, the achievement of temporary lanes closing, and clearance [5][6][7].…”

Section: Literature Surveymentioning

confidence: 99%

“…Considering that CM and ALR design approaches have different operating ways, there will also be different physical design elements, and such differences are highlighted in Table 1. Despite considerations about managed lane advantages, a different perspective must be studied regarding geometric characteristics and traffic control devices [7,8]. Permanent or temporary conversion of the hard shoulder into an ordinary running lane in terms of benefits causes a significant capacity increase; however, a disadvantage that must be taken into account is characterized by a safety reduction because of the hard shoulder removal.…”

mentioning

confidence: 99%

Managed Lane as Strategy for Traffic Flow and Safety: A Case Study of Catania Ring Road

et al. 2022

View full text Add to dashboard Cite

According to international traffic statistics, the quantity of daily traveling vehicles amounts to one billion, and it is estimated that in 2050 that amount will reach four billion. Urban population is growing especially in large metropolitan areas. This increase has different outgrowths concerning regular circulation, safety, and climatic perspectives. Therefore, the present and further increase of traffic flow is affecting the operational and safety performance of several roadway categories. Urban ring roads belong to a roadway category that is particularly subjected to congestion phenomena, but in many cases, public administrations are unable to afford required expenses to upgrade or build new roadways. Therefore, an intervention is needed that increases the capacity of existing infrastructures with limited cost results to be fundamental. In such a framework, the dynamic use of hard shoulder running (HSR) would represent a valid solution that both exploits existing infrastructures and facilitates traffic outflow by implementation of smart digital roads, with limited interventions on the physical infrastructure. Despite the expected improvements in traffic capacity, the HSR poses safety issues particularly in specific locations (e.g., interchanges) and for the operation of the transition phase for opening and closing the HSR. In this paper, the applicability and usefulness of microsimulation were tested with the Catania (Italy) ring road, as a case study, through the application of VISSIM traffic microsimulation software and SSAM traffic conflict tool, implementing different scenarios for the activation of the HSR.

show abstract

“…Semi‐supervised learning includes expectation maximization (EM) algorithms and multi‐view learning. Thanks to computing advances, ML is being widely tested across various areas of transport data analysis, including predictions of truck‐type category (Huang & Kockelman, 2020), mode type (Pirra & Diana, 2019; Wang & Zhao, 2020; Xie et al., 2003), travel times (Zhang & Haghani, 2015), injury severity (Das et al., 2019; Delen et al., 2017; Hamad et al., 2019), traffic flow (Cui et al., 2019; Hosseini & Talebpour, 2019), trip purpose (Deng & Ji, 2010), electric vehicle‐charging schedules (Chung et al., 2019; Jahangir et al., 2019), automated‐vehicle applications—like object identification and driving response (Liang & Wang, 2021), and traveler choices of managed (and tolled) lanes (Ashraf et al., 2021).…”

Section: Introductionmentioning

confidence: 99%

How does machine learning compare to conventional econometrics for transport data sets? A test of ML versus MLE

Kockelman

2021

Growth and Change

View full text Add to dashboard Cite

Machine learning (ML) is being used regularly in many different fields. This paper compares traditional econometric methods that have better explanations of data analysis to ML methods, focusing on predicting, understanding and unpacking ML methods which have higher prediction accuracies of four key transport‐planning variables: household vehicle‐miles traveled (continuous variable), household vehicle ownership (count variable), mode choice (categorical variable), and land use change (categorical variable with strong spatial interactions). Here, the results of ten ML methods are compared to methods of ordinary least squares (OLS), multinomial logit (MNL), negative binomial and spatial auto‐regressive (SAR). The U.S.’s 2017 National Household Travel Survey and land use data sets from the Dallas‐Ft. Worth region of Texas are used. Results suggest traditional econometric methods work pretty well on the more continuous responses (VMT and vehicle ownership), but the random forest (RF), gradient boosting decision trees (GBDT), and extreme gradient boosting (XGBoost) methods delivered the best results, though the RF model required 30 to almost 60 times more computing time than XGBoost and GBDT methods. The RF, GBDT, XGBoost, light gradient boosting method (lightGBM), and catboost offer better results than other methods for the two “classification” cases, with lightGBM being the most time‐efficient. Importantly, ML methods captured the plateauing effect modelers may expect when extrapolating covariate effects.

show abstract

Predicting the Use of Managed Lanes Using Machine Learning

Cited by 3 publications

References 24 publications

Using Behavioral Economics to Identify Potential Managed Lane Users

Using Behavioral Economics to Identify Potential Managed Lane Users

Managed Lane as Strategy for Traffic Flow and Safety: A Case Study of Catania Ring Road

How does machine learning compare to conventional econometrics for transport data sets? A test of ML versus MLE

Contact Info

Product

Resources

About