Article: Multivariate short-term traffic flow prediction based on real-time expressway toll plaza data using non-parametric techniques Journal: International Journal of Vehicle Information and Communication Systems (IJVICS) 2022 Vol.7 No.1 pp.32 - 50 Abstract: Accurate real-time traffic flow prediction is a vital component of Intelligent Transportation System (ITS). The real-time traffic flow prediction helps transportation authorities as well as travellers for better route guidance. In this study, a novel approach is proposed for accurate Toll Plaza traffic prediction by introducing heterogeneous data sources other than traffic volume data. Toll data is analysed with exogenous factors such as weather conditions and holidays. Here, ten non-parametric techniques are applied for traffic prediction on real-time multivariate data set. The proposed approach is validated using data collected from Pinjore-Kalka Toll Plaza, Chandigarh, India. The performances of the non-parametric models are compared on the basis of mean square error, absolute mean square error, coefficient of determination and correlation. The experimental results revealed that random forest regression technique outperforms other techniques and achieved accuracy 90%. The proposed approach can be used for further proxy measure of Level of Service (LOS) to design the existing infrastructure more efficiently for the application purpose in smart cities. Inderscience Publishers - linking academia, business and industry through research

Title: Multivariate short-term traffic flow prediction based on real-time expressway toll plaza data using non-parametric techniques

Authors: Annu Mor; Mukesh Kumar; Santanu Chaudhury

Addresses: University Institute of Engineering and Technology, Panjab University, Chandigarh, Punjab, India ' University Institute of Engineering and Technology, Panjab University, Chandigarh, Punjab, India ' Indian Institute of Technology-Jodhpur (IIT-Jodhpur), Rajasthan, India

Abstract: Accurate real-time traffic flow prediction is a vital component of Intelligent Transportation System (ITS). The real-time traffic flow prediction helps transportation authorities as well as travellers for better route guidance. In this study, a novel approach is proposed for accurate Toll Plaza traffic prediction by introducing heterogeneous data sources other than traffic volume data. Toll data is analysed with exogenous factors such as weather conditions and holidays. Here, ten non-parametric techniques are applied for traffic prediction on real-time multivariate data set. The proposed approach is validated using data collected from Pinjore-Kalka Toll Plaza, Chandigarh, India. The performances of the non-parametric models are compared on the basis of mean square error, absolute mean square error, coefficient of determination and correlation. The experimental results revealed that random forest regression technique outperforms other techniques and achieved accuracy 90%. The proposed approach can be used for further proxy measure of Level of Service (LOS) to design the existing infrastructure more efficiently for the application purpose in smart cities.

Keywords: traffic flow; ITS; intelligent transportation system; non-parametric technique; multivariate time series data set; proxy measure level of service.

DOI: 10.1504/IJVICS.2022.120821

International Journal of Vehicle Information and Communication Systems, 2022 Vol.7 No.1, pp.32 - 50

Received: 25 Sep 2019
Accepted: 08 Apr 2020
Published online: 11 Feb 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Multivariate short-term traffic flow prediction based on real-time expressway toll plaza data using non-parametric techniques

Keep up-to-date