Estimating Freeway LevelofService Using Crowdsourced Data
Abstract
:1. Introduction
2. Literature Review
2.1. Traffic Status and LOS Assessment Methods
2.2. Travel Time Reliability
2.3. Alternative LOS Methods
2.4. Waze Data
2.5. Gap in the Literature
3. Data
3.1. Waze Speed and Travel Time Data
3.2. Waze Crowdsourced Alert Data
3.3. Fixed Location Data
3.4. Study Time and Area
4. Methodology
 Step 1: Data collection, which includes archiving Waze data and traditional fixed location sensor data, as well as preprocessing and normalization;
 Step 2: Extract model inputs, which includes statistical measures, travel time performance measures, and crowdsourced Waze alerts;
 Step 3: Calculating ground truth LOS, using fixed location sensors, and labeling observations of Waze input data with the corresponding ground truth data;
 Step 4: LOS assessment, by performing different machine learning methods. This part includes feature selection, cross validation, and selecting the preferred method.
4.1. Step 1: Data Collection
4.2. Step 2: Model Inputs
 Basic statistical measures, including the average speed, standard deviation, range, coefficient of variation, standard error, percentiles (25th, 50th, and 90th), and interquartile range.
 Travel time performance measures, including the Travel Time Index, Planning Time Index, and Buffer Time Index.
 Crowdsourced data, including the number of users’ accident, jam, and hazard reports in the Waze alerts data.
4.2.1. Basic Statistical Measures
4.2.2. Travel Time Performance Measures
 The Travel Time Index (TTI) captures the travel time variation by calculating the average travel time ratio to the free flow travel time in the segment. This index explores how the travel time deviates from the free flow travel time during the intended period, which is typically LOS A [30].
4.2.3. Crowdsourced Data
4.3. Step 3: Ground Truth LOS
4.4. Step 4: Machine Learning Methods
 Random Forest (RF): RF is an ensemble classification method that combines several random decision trees. In this method, all trees are built independently. Then, it classifies the data based on the majority of votes of all trees;
 Support Vector Machines (SVM): SVMs are wellknown marginbased classification methods. For each class, the SVM algorithm finds the optimal support vector that provides the maximum distance to other classes. By calculating the optimal support vectors, the algorithm can identify the boundaries and classify the data;
 KNearest Neighbor (KNN): KNNs are nonparametric methods that are widely used for classification. All training data are considered in an ndimensional feature space (n = number of input features) in this method. For each observation, the algorithm looks for the k (a predefined constant) nearest neighbors based on the Euclidean distance. Then, it assigns the category based on the most frequent label of the neighbors.
5. Results
5.1. Descriptive Statistics
5.2. LOS Classification Model Using Machine Learning
5.2.1. Model Training and Hyperparameter Tuning
 RF:
 
 Number of trees: A higher number of trees typically avoids overfitting;
 
 Maximum number of features: A smaller number of features basically reduces the chance of overfitting;
 
 Maximum tree depth: The lower the tree depth, the less likely overfitting is;
 KNN:
 
 Number of neighbors (K): Increasing the number of neighbors can avoid overfitting;
 SVM:
 
 C: Demonstrates a tradeoff between a high and low accuracy, with a low C value resulting in a smoother decision surface and a lower chance of overfitting.
 
 Sigma: A large gamma value can cause overfitting.
5.2.2. Model Selection
 Model I uses only travel time performance measures as the model inputs and shows how accurately travel time performance measures can determine LOS;
 Model II uses travel time performance measures and basic statistical measures as the inputs;
 Model III incorporates crowdsourced Waze alerts and uses all three types of input. Model III captures the impact of the crowdsourced alerts in terms of improvement of the LOS classification.
5.2.3. Test Result
5.3. Sensitivity Analysis
5.4. Variable Importance
6. Limitations and Future Work
7. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
 Board, T.R. Highway Capacity Manual 6th Edition: A Guide for Multimodal Mobility Analysis; The National Academies Press: Washington, DC, USA, 2016; p. 6. [Google Scholar]
 Khan, S.M.; Dey, K.C.; Chowdhury, M. Realtime traffic state estimation with connected vehicles. Ieee Trans. Intell. Transp. Syst. 2017, 18, 1687–1699. [Google Scholar] [CrossRef]
 Hernandez, S.; Tok, A.; Ritchie, S.G. Density Estimation Using Inductive Loop Signature Based Vehicle ReIdentification and Classification; Institute of Transportation Studies University of California: Irvine, CA, USA, 2013. [Google Scholar]
 Hargrove, S.R.; Lim, H.; Han, L.D.; Freeze, P.B. Empirical evaluation of the accuracy of technologies for measuring average speed in real time. Transp. Res. Rec. 2016, 2594, 73–82. [Google Scholar] [CrossRef]
 Underwood, S.E. A Review and Classification of Sensors for Intelligent VehicleHighway Systems; University of Michigan Transportation Research Institute: Ann Arbor, MI, USA, 1990. [Google Scholar]
 Zhang, J.; Wang, F.Y.; Wang, K.; Lin, W.H.; Xu, X.; Chen, C. Datadriven intelligent transportation systems: A survey. IEEE Trans. Intell. Transp. Syst. 2011, 12, 1624–1639. [Google Scholar] [CrossRef]
 Herrera, J.C.; Work, D.B.; Herring, R.; Ban, X.J.; Jacobson, Q.; Bayen, A.M. Evaluation of traffic data obtained via GPSenabled mobile phones: The Mobile Century field experiment. Transp. Res. Part C Emerg. Technol. 2010, 18, 568–583. [Google Scholar] [CrossRef][Green Version]
 Hoseinzadeh, N.; Arvin, R.; Khattak, A.J.; Han, L.D. Integrating safety and mobility for pathfinding using big data generated by connected vehicles. J. Intell. Transp. Syst. 2020, 24, 404–420. [Google Scholar] [CrossRef]
 Khattak, A.J.; Mahdinia, I.; Mohammadi, S.; Mohammadnazar, A.; Wali, B. Big Data Generated by Connected and Automated Vehicles for Safety Monitoring, Assessment and Improvement, Final Report (Year 3). arXiv 2021, arXiv:06106. [Google Scholar]
 Mohammadnazar, A.; Arvin, R.; Khattak, A.J. Classifying travelers’ driving style using basic safety messages generated by connected vehicles: Application of unsupervised machine learning. Transp. Res. Part C Emerg. Technol. 2021, 122, 102917. [Google Scholar] [CrossRef]
 Mousavi, S.M.; Lord, D.; Dadashova, B.; Mousavi, S.R. Can Autonomous Vehicles Enhance Traffic Safety at Unsignalized Intersections? In Proceedings of the International Conference on Transportation and Development, Seattle, WA, USA, 26–29 May 2020; American Society of Civil Engineers: Reston, VA, USA, 2020; pp. 194–206. [Google Scholar]
 Mousavi, S.M.; Osman, O.A.; Lord, D.; Dixon, K.K.; Dadashova, B. Investigating the safety and operational benefits of mixed traffic environments with different automated vehicle market penetration rates in the proximity of a driveway on an urban arterial. Accid. Anal. Prev. 2021, 152, 105982. [Google Scholar] [CrossRef] [PubMed]
 Azad, M.; Hoseinzadeh, N.; Brakewood, C.; Cherry, C.R.; Han, L.D. Fully autonomous buses: A literature review and future research directions. J. Adv. Transp. 2019, 2019, 4603548. [Google Scholar] [CrossRef][Green Version]
 Azad, M.; Hoseinzadeh, N.; Brakewood, C.; Cherry, C.R.; Han, L.D. A literature review on fully autonomous buses. In Proceedings of the Transportation Research Board 98th Annual 2019, Washington, DC, USA, 13–17 January 2019. [Google Scholar]
 Mahdinia, I.; Mohammadnazar, A.; Arvin, R.; Khattak, A.J. Integration of automated vehicles in mixed traffic: Evaluating changes in performance of following humandriven vehicles. Accid. Anal. Prev. 2021, 152, 106006. [Google Scholar] [CrossRef]
 Mahdinia, I.; Arvin, R.; Khattak, A.J.; Ghiasi, A. Safety, energy, and emissions impacts of adaptive cruise control and cooperative adaptive cruise control. Transp. Res. Rec. 2020, 2674, 253–267. [Google Scholar] [CrossRef]
 Chatzimilioudis, G.; Konstantinidis, A.; Laoudias, C.; ZeinalipourYazti, D. Crowdsourcing with smartphones. IEEE Internet Comput. 2012, 16, 36–44. [Google Scholar] [CrossRef]
 Kanhere, S.S. Participatory sensing: Crowdsourcing data from mobile smartphones in urban spaces. In Proceedings of the 2011 IEEE 12th International Conference on Mobile Data Management, Luleå, Sweden, 6–9 June 2011; Volume 2, pp. 3–6. [Google Scholar]
 Nair, D.J.; Gilles, F.; Chand, S.; Saxena, N.; Dixit, V. Characterizing multicity urban traffic conditions using crowdsourced data. PLoS ONE 2019, 14, e0212845. [Google Scholar]
 Pack, M.; Ivanov, N. Are You Gonna Go My WAZE? Inst. Transp. Eng. ITE J. 2017, 87, 28. [Google Scholar]
 Hoseinzadeh, N.; Liu, Y.; Han, L.D.; Brakewood, C.; Mohammadnazar, A. Quality of locationbased crowdsourced speed data on surface streets: A case study of Waze and Bluetooth speed data in Sevierville, TN. Comput. Environ. Urban. Syst. 2020, 83, 101518. [Google Scholar] [CrossRef]
 Senarath, Y.; Nannapaneni, S.; Purohit, H.; Dubey, A. Emergency Incident Detection from Crowdsourced Waze Data using Bayesian Information Fusion. arXiv 2020, arXiv:05440. [Google Scholar]
 Ali, F.; Ali, A.; Imran, M.; Naqvi, R.A.; Siddiqi, M.H.; Kwak, K.S. Traffic accident detection and condition analysis based on social networking data. Accid. Anal. Prev. 2021, 151, 105973. [Google Scholar] [CrossRef]
 Farajiparvar, P.; Hoseinzadeh, N.; Han, L.D.; Hedayatipour, A. Deep Learning Techniques for Traffic Speed Forecasting with Side Information. In Proceedings of the 2020 IEEE Green Energy and Smart Systems Conference (IGESSC), Long Beach, CA, USA, 2–3 November 2020; pp. 1–5. [Google Scholar]
 Li, X.; Dadashova, B.; Yu, S.; Zhang, Z. Rethinking Highway Safety Analysis by Leveraging Crowdsourced Waze Data. Sustainability 2020, 12, 10127. [Google Scholar] [CrossRef]
 Chen, Y.; Zhou, M.; Zheng, Z.; Huo, M. Toward practical crowdsourcingbased road anomaly detection with scaleinvariant feature. IEEE Access 2019, 7, 67666–67678. [Google Scholar] [CrossRef]
 Daraghmi, Y.A.; Wu, T.H. CrowdsourcingBased Road Surface Evaluation and Indexing. IEEE Trans. Intell. Transp. Syst. 2020. [Google Scholar] [CrossRef]
 Martinez, M.; Yang, K.; Constantinescu, A.; Stiefelhagen, R.J.S. Helping the Blind to Get through COVID19: Social Distancing Assistant Using RealTime Semantic Segmentation on RGBD Video. Sensors 2020, 20, 5202. [Google Scholar] [CrossRef] [PubMed]
 AminNaseri, M.; Chakraborty, P.; Sharma, A.; Gilbert, S.B.; Hong, M. Evaluating the reliability, coverage, and added value of crowdsourced traffic incident reports from Waze. Transp. Res. Rec. 2018, 2672, 34–43. [Google Scholar] [CrossRef][Green Version]
 Singh, V.; Gore, N.; Chepuri, A.; Arkatkar, S.; Joshi, G.; Pulugurtha, S. Examining Travel Time Variability and Reliability on an Urban Arterial Road Using WiFi Detections—A Case Study. J. East. Asia Soc. Transp. Stud. 2019, 13, 2390–2411. [Google Scholar]
 Celikoglu, H.B. An approach to dynamic classification of traffic flow patterns. Comput. Aided Civ. Infrastruct. Eng. 2013, 28, 273–288. [Google Scholar] [CrossRef]
 Sekuła, P.; Marković, N.; Laan, Z.V.; Sadabadi, K.F. Estimating historical hourly traffic volumes via machine learning and vehicle probe data: A Maryland case study. Transp. Res. Part C Emerg. Technol. 2018, 97, 147–158. [Google Scholar] [CrossRef][Green Version]
 Marwah, B.; Singh, B. Level of service classification for urban heterogeneous traffic: A case study of Kanpur metropolis. In Proceedings of the Fourth International Symposium on Highway Capacity, Maui, HI, USA, 27 June–1 July 2000. [Google Scholar]
 Altintasi, O.; TuydesYaman, H.; Tuncay, K. Detection of urban traffic patterns from Floating Car Data (FCD). Transp. Res. Procedia 2017, 22, 382–391. [Google Scholar] [CrossRef]
 Geistefeldt, J.; Giuliani, S.; Vortisch, P.; Leyn, U.; Trapp, R.; Busch, F.; Rascher, A.; Celikkaya, N. Assessment of level of service on freeways by microscopic traffic simulation. Transp. Res. Rec. 2014, 2461, 41–49. [Google Scholar] [CrossRef]
 Wu, N.; Lemke, K. A new model for level of service of freeway merge, diverge, and weaving segments. ProcediaSoc. Behav. Sci. 2011, 16, 151–161. [Google Scholar] [CrossRef]
 Jolovic, D.; Stevanovic, A.; Sajjadi, S.; Martin, P.T. Assessment of LevelOfService for Freeway Segments Using HCM and Microsimulation Methods. Transp. Res. Procedia 2016, 15, 403–416. [Google Scholar] [CrossRef][Green Version]
 Celikoglu, H.B.; Silgu, M.A. Extension of traffic flow pattern dynamic classification by a macroscopic model using multivariate clustering. Transp. Sci. 2016, 50, 966–981. [Google Scholar] [CrossRef]
 Aljamal, M.A.; Abdelghaffar, H.M.; Rakha, H.A. Developing a Neural–Kalman Filtering Approach for Estimating Traffic Stream Density Using Probe Vehicle Data. Sensors 2019, 19, 4325. [Google Scholar] [CrossRef] [PubMed][Green Version]
 Wassantachat, T.; Li, Z.; Chen, J.; Wang, Y.; Tan, E. Traffic density estimation with online SVM classifier. In Proceedings of the 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance, Genoa, Italy, 2–4 September 2009; pp. 13–18. [Google Scholar]
 Khan, S.M. RealTime Traffic Condition Assessment with Connected Vehicles. Master’s Thesis, Clemson University, Clemson, SC, USA, 2015. [Google Scholar]
 Kurniawan, F.; Sajati, H.; Dinaryanto, O. Image Processing Technique for Traffic Density Estimation. Int. J. Eng. Technol. 2017, 9, 1496–1503. [Google Scholar] [CrossRef][Green Version]
 Lomax, T.; Margiotta, R. Selecting Travel Reliability Measures; Citeseer: University Park, TX, USA, 2003. [Google Scholar]
 Schrank, D.; Eisele, B.; Lomax, T. TTI’s 2012 Urban Mobility Report; Texas A&M Transportation Institute, The Texas A&M University System: College Station, TX, USA, 2012; Volume 4. [Google Scholar]
 Martchouk, M.; Mannering, F.L.; Singh, L. Travel time reliability in Indiana; Publication FHWA/IN/JTRP2010/08; Joint Transportation Research Program, Indiana Department of Transportation and Purdue University: West Lafayette, IN, USA, 2010. [Google Scholar] [CrossRef][Green Version]
 Chepuri, A.; Wagh, A.; Arkatkar, S.S.; Joshi, G. Study of travel time variability using twowheeler probe data–an Indian experience. In Proceedings of the Institution of Civil EngineersTransport, London, UK, 4 August 2018; Thomas Telford Ltd.: London, UK, 2018; Volume 171, pp. 190–206. [Google Scholar] [CrossRef]
 Kittelson, W.; Vandehey, M. Incorporation of Travel Time Reliability into the HCM; Transportation Research Board: Washington, DC, USA, 2013. [Google Scholar]
 Chen, C.; Skabardonis, A.; Varaiya, P. Traveltime reliability as a measure of service. Transp. Res. Rec. 2003, 1855, 74–79. [Google Scholar] [CrossRef][Green Version]
 Pulugurtha, S.S.; Imran, M.S. Modeling basic freeway section levelofservice based on travel time and reliability. Case Stud. Transp. Policy 2020, 8, 127–134. [Google Scholar] [CrossRef]
 Kodupuganti, S.R.; Pulugurtha, S.S. Linklevel travel time measuresbased level of service thresholds by the posted speed limit. Transport. Res. Interdiscipl. Perspect. 2019, 3, 100068. [Google Scholar] [CrossRef]
 Misra, A.; Gooze, A.; Watkins, K.; Asad, M.; Le Dantec, C.A. Crowdsourcing and its application to transportation data collection and management. Transp. Res. Rec. 2014, 2414, 1–8. [Google Scholar] [CrossRef][Green Version]
 Santos, S.R.d.; Davis, C.A., Jr.; Smarzaro, R. Integration of data sources on traffic accidents. In Proceedings of the GeoInfo, Campos do Jordão, Brazil, 27–30 November 2016; pp. 192–203. [Google Scholar]
 Bahaweres, R.B.; Akbar, M.R. Analysis of travel time computation accuracy from Crowdsource data of hospitality application in South of Tangerang City with estimated travel time method. In Proceedings of the 2017 5th International Conference on Cyber and IT Service Management (CITSM), Bali, Indonesia, 8–10 August 2017; pp. 1–5. [Google Scholar]
 Turner, S.; Martin, M.; Griffin, G.; Le, M.; Das, S.; Wang, R.; Dadashova, B.; Li, X. Exploring Crowdsourced Monitoring Data for Safety; Bureau of Transportation Statistics U.S. Department of Transportation: Washington, DC, USA, 2020.
 Perez, G.V.A.; Lopez, J.C.; Cabello, A.L.R.; Grajales, E.B.; Espinosa, A.P.; Fabian, J.L.Q. Road Traffic Accidents Analysis in Mexico City through Crowdsourcing Data and Data Mining Techniques. Int. J. Comput. Inf. Eng. 2018, 12, 604–608. [Google Scholar]
 Sanchez, R.; Martinez, D.; Mitnik, O.A.; YanezPagans, P.; Lanzalot, M.L.; Stucchi, R.; Sanguino, L. Dynamic Traffic Lights and Urban Mobility: An Application of Waze Data to the City of Medellın; The Research Institute for Development, Growth and Economics (RIDGE): Montevideo, Uruguay, 2019. [Google Scholar]
 Naming the Coronavirus Disease (COVID19) and the Virus that Causes it. 2020. Available online: https://www.who.int/emergencies/diseases/novelcoronavirus2019/technicalguidance/namingthecoronavirusdisease(covid2019)andthevirusthatcausesit/ (accessed on 8 February 2021).
 AlSobky, A.S.A.; Mousab, R.M. Traffic density determination and its applications using smartphone. Alex. Eng. J. 2016, 55, 513–523. [Google Scholar] [CrossRef][Green Version]
 Ensley, J.O. Application of Highway Capacity Manual 2010 LevelofService Methodologies for Planning Deficiency Analysis. Master’s Thesis, University of Tennessee, Knoxville, TN, USA, 2012. [Google Scholar]
No.  Reference  Year  Data  Index Used  Method 

1  Kittelson et al. [47]  2008  Sensor data (speed, density, travel time) 


2  Altinatasi et al. [34].  2016  Floating Car Data (speed) 


3  Khan, Dey, and Chowdhury [2]  2017  Simulation (speed, density) 


4  Singh et al. [30]  2019  WiFi probe vehicle (speed, travel time) 


5  Kodupuganti, and Pulugurtha [50]  2019  Travel time data provided by North Carolina DOT 


6  Pulugurtha and Imran [49]  2020  Simulation (travel time) 


No.  Reference  Year  Country  Waze Data Used  Findings and Application 

1  Santos et al. [52]  2016  Brazil  Waze accident alerts 

2  Bahaweres et al. [53]  2017  Indonesia  Waze travel time 

3  Pack and Ivanov [20]  2017  United States  Waze alerts 

4  AminNaseri et al. [29]  2018  United States  Waze congestion and accident alerts 

5  Perez et al. [55]  2018  Mexico  Waze alerts 

6  Raul Sanchez et al. [56]  2019  Colombia  Waze jam and accident alerts 

7  Turner et al. [54]  2020  United States  Waze alerts 

8  Hoseinzadeh et al. [21]  2020  United States  Waze speed 

9  Li et al. [25]  2020  United States  Waze alerts 

10  Senarath et al. [22]  2020  United States  Waze alerts 

Model Input  Measure  Equation  Eq. No. 

Basic statistical measures of speed  Average Speed  $\overline{v}=\frac{{{\displaystyle \sum}}_{1}^{n}{v}_{i}}{n}$ $\mathrm{where}{v}_{i}\text{}\mathrm{is}\text{}\mathrm{the}\text{}\mathrm{speed}\text{}\mathrm{and}n\mathrm{is}\text{}\mathrm{the}\text{}\mathrm{number}$ $\mathrm{of}\text{}\mathrm{observations}\text{}\mathrm{in}\text{}\mathrm{each}\text{}\mathrm{time}\text{}\mathrm{interval}$  $\left(1\right)$ 
Standard Deviation (SD)  $\sigma =\sqrt{\frac{{{\displaystyle \sum}}_{1}^{n}{({v}_{i}\overline{v})}^{2}}{n}}$  $\left(2\right)$  
Range  $Range\left(v\right)=\underset{i=1,2,\dots ,n}{\mathrm{max}}{v}_{i}\underset{i=1,2,\dots ,n}{\mathrm{min}}{v}_{i}$  $\left(3\right)$  
Coefficient of Variation (CoV)  $CoV=\frac{\sigma}{\overline{v}}$  $\left(4\right)$  
Standard Error (SE)  $SE=\frac{\sigma}{\sqrt{n}}$  $\left(5\right)$  
Percentiles (25th, 50th, 75th, 90th)  $kthpercentile=rank\left(\frac{k}{100}\left(n+1\right)\right)$ $\mathrm{where}k=25,50,75,90$ Here, rank is ordering the dataset from smallest to largest and finds the value with the $\frac{k}{100}\left(n+1\right)$ index  $\left(6\right)$  
Interquartile Range (IQR)  $IQR={Q}_{3}{Q}_{1}$ $\mathrm{where}{Q}_{3}\mathrm{is}\text{}\mathrm{the}\text{}75\mathrm{th}\text{}\mathrm{percentile}\text{}\mathrm{and}$ ${Q}_{1}\text{}\mathrm{is}\text{}\mathrm{the}\text{}25\mathrm{th}\text{}\mathrm{percentile}\text{}\mathrm{of}{v}_{i}$  $\left(7\right)$  
Travel time performance  Travel Time Index (TTI)  $TTI=\frac{T{T}_{Avg}}{T{T}_{freeflow}}$ $\mathrm{where}T{T}_{Avg}\mathrm{is}\text{}\mathrm{the}\text{}\mathrm{average}\text{}\mathrm{travel}\text{}\mathrm{time}\text{}\mathrm{and}$ $T{T}_{freeflow}\text{}\mathrm{is}\text{}\mathrm{the}\text{}\mathrm{free}\text{}\mathrm{flow}\text{}\mathrm{travel}\text{}\mathrm{time}$  (8) 
Buffer Time Index (BTI)  $BTI=\frac{T{T}_{95th}T{T}_{Avg}}{T{T}_{Avg}}$ $\mathrm{where}\text{}T{T}_{95th}\text{}\mathrm{is}\text{}\mathrm{the}\text{}95\mathrm{th}\text{}\mathrm{percentile}\text{}\mathrm{of}\text{}\mathrm{the}\text{}\mathrm{travel}\text{}\mathrm{time}$  (9)  
Planning Time Index (PTI)  $PTI=\frac{T{T}_{95th}}{T{T}_{freeflow}}$  (10)  
Crowdsourced data  Hourly Number of Alerts  $Count(WazeAler{t}_{t}^{s}$) $\mathrm{where}s\text{}\mathrm{is}\text{}\mathrm{the}\text{}\mathrm{study}\text{}\mathrm{segment}\text{}\mathrm{and}t\mathrm{is}$ $\mathrm{the}\text{}\mathrm{time}\text{}\mathrm{intervel}\text{}\left(\mathrm{hour}\text{}\mathrm{of}\text{}\mathrm{day}\right)$  (11) 
LOS  Density (Vehicle/Mile/Lane)  Description 

A  ≤11  Free flow 
B  >11–18  Reasonably free flow 
C  >18–26  Stable flow (acceptable delays) 
D  >26–35  Speeds decline slightly with increasing flows 
E  >35–45  Operation near or at capacity 
F  >45  Forced or breakdown flow 
Model Input  Measure  Mean  Min.  Max.  Median  S.D. 

Basic statistical measures on speed  Average speed (km/h)  100.6  31.1  119.4  110.7  21.7 
Speed standard deviation (km/h)  6.6  0.0  39.4  4.0  6.3  
Minimum speed (km/h)  88.4  18.0  118.4  105.4  25.7  
Maximum speed (km/h)  111.0  39.4  146.0  118.4  20.1  
Range of speed (km/h)  22.7  0.0  91.1  14.5  18.2  
CoV of speed  0.0  0.0  0.5  0.0  0.1  
SE of speed  0.5  0.0  3.1  0.3  0.5  
25th percentile (km/h)  96.9  25.9  114.4  109.8  24.5  
50th percentile (km/h)  100.9  29.1  118.4  110.7  22.9  
75th percentile (km/h)  105.1  36.5  121.0  111.5  21.1  
90th percentile (km/h)  107.7  39.4  126.5  118.4  20.3  
Travel time performance  IQR  5.1  0.0  51.9  3.8  7.7 
TTI  1.2  1.0  3.9  1.0  0.5  
BTI  0.2  −0.4  3.4  0.0  0.4  
PTI  1.4  1.0  5.8  1.1  0.7  
Crowdsourced data  Number of Waze alerts  9.0  0.0  101.0  4.0  20.0 
Classifier  3Fold Cross Validation  5Fold Cross Validation  10Fold Cross Validation  

Accuracy  Kappa  Accuracy  Kappa  Accuracy  Kappa  
SVM  0.91  0.81  0.91  0.81  0.90  0.79 
RF  0.91  0.82  0.93  0.83  0.92  0.83 
KNN  0.88  0.77  0.89  0.79  0.88  0.76 
Date  Random Forest Test Result  

Accuracy  Kappa  
03/15/2020 to 04/15/2020  0.95  0.86 
08/01/2020 to 08/31/2020  0.92  0.83 
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. 
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Hoseinzadeh, N.; Gu, Y.; Han, L.D.; Brakewood, C.; Freeze, P.B. Estimating Freeway LevelofService Using Crowdsourced Data. Informatics 2021, 8, 17. https://doi.org/10.3390/informatics8010017
Hoseinzadeh N, Gu Y, Han LD, Brakewood C, Freeze PB. Estimating Freeway LevelofService Using Crowdsourced Data. Informatics. 2021; 8(1):17. https://doi.org/10.3390/informatics8010017
Chicago/Turabian StyleHoseinzadeh, Nima, Yangsong Gu, Lee D. Han, Candace Brakewood, and Phillip B. Freeze. 2021. "Estimating Freeway LevelofService Using Crowdsourced Data" Informatics 8, no. 1: 17. https://doi.org/10.3390/informatics8010017