Prediction of Average Annual Daily Traffic Using Machine Learning Methods

Prediction of Average Annual Daily Traffic Using Machine Learning Methods

International Journal of Computer Science and Engineering
© 2019 by SSRG - IJCSE Journal
Volume 6 Issue 11
Year of Publication : 2019
Authors : Srinivasan Suresh
: 10.14445/23488387/IJCSE-V6I11P111

pdf
Citation:
MLA Style:

Srinivasan Suresh, "Prediction of Average Annual Daily Traffic Using Machine Learning Methods" SSRG International Journal of Computer Science and Engineering 6.11 (2019): 51-54.

APA Style:

Srinivasan Suresh,(2019). Prediction of Average Annual Daily Traffic Using Machine Learning Methods. SSRG International Journal of Computer Science and Engineering 6(11), 51-54.

Abstract:

The field of machine learning is growing enormously and been widely used. There are many techniques been developed in the machine learning arena. Using these techniques to predict the Average Annual Daily Traffic (AADT) would help to improve the accuracy in prediction and to plan routes in a better manner. The Linear regression, Ridge regression and Lasso regression methods are used in this analysis. The data is obtained from the Highway Performance Management System (HPMS) and from Virginia Roads, the official data provider for the State of Virginia. The raw data is cleansed, regrouped and prepared for the regression analysis. Route category, Through lanes, Facility type and Functional class are the key variables used in the analysis. The machine learning models built through these methods have almost 77% accuracy and these models can be reused to predict the AADT values for new routes or extension of routes.

References:

[1] Oleszak, M. (2019, November 12). Regularization: Ridge, Lasso and Elastic Net. Retrieved from https://www.datacamp.com/community/tutorials/tutorial-ridgelasso-elastic-net
[2] Annapoorani Anantharaman,(2019). A Study of Logistic Regression And Its Optimization Techniques Using Octave. SSRG International Journal of Computer Science and Engineering 6(10), 23-28.
[3] Selby, B. &Kockelman, K. M. (2011, January). Spatial prediction of AADT in unmeasured locations by universal kriging.Retrieved from https://www.caee.utexas.edu/prof/kockelman/public_html/TRB11AADTUnivKriging.pdf
[4] Sliupas, T. (2006). Annual average daily traffic forecasting using different techniques. Transport, 21:1, 38-43, doi: 10.1080/16484142.2006.9638039
[5] U. S. Department of Transportation (2018). Traffic Data Computation Method (FHWA-PL-18-027). Retrieved from
[6] https://www.fhwa.dot.gov/policyinformation/pubs/pl18027_traffic_data_pocket_guide.pdf
[7] U. S. Department of Transportation (2018). HPMS Public Release of Geospatial Data in Shapefile Format. Retrievedfrom https://www.fhwa.dot.gov/policyinformation/hpms/shapefiles.cfm
[8] U. S. Department of Transportation (2019). Cost Effective Strategies for Estimating Statewide AADT (FHWA-SC-18-10). Retrieved from https://www.scdot.scltap.org/wpcontent/uploads/2019/04/SPR-717-Final-Report.pdf
[9] Virginia Department of Transportation (2019). LRS Route Overlap. Retrieved from https://www.virginiaroads.org/datasets/lrs-route-overlap
[10] Wu, J. & Xu, H. (2019). Annual Average Daily Traffic Prediction Model for Minor Roads at Intersections. Journal of Transportation Engineering, 145(10),doi:10.1061/JTEPBS.0000262
[11] Peixerio, M. (2019, January 12). How to Perform Lasso and Ridge Regression in Python. Retrieved from https://towardsdatascience.com/how-to-perform-lasso-andridge-regression-in-python-3b3b75541ad8
[12] Regression Visualizers. (n. d.). Retrieved from https://www.scikit-yb.org/en/latest/api/regressor/
[13] Shakti Chourasiya, Suvrat Jain,(2019). A Study Review On Supervised Machine Learning Algorithms. SSRG International Journal of Computer Science and Engineering 6(8), 16-2

Key Words:

Annual Average Daily Traffic (AADT), Machine Learning, Through Lanes, Functional System