Title: Traffic time series forecasting on highways - a contemporary survey of models, methods and techniques

Authors: G. Jayanthi; P. Jothilakshmi

Addresses: Faculty of Information and Communication Engineering, Sri Venkateswara College of Engineering, Anna University (Chennai), Sriperumbudur, Tamil Nadu, India ' Department of Electronics and Communication Engineering, Sri Venkateswara College of Engineering, Sriperumbudur, Tamil Nadu, India

Abstract: Transportation research is dynamic and essential engineering prospect of all nations across the globe. Recent developments in intelligent transport systems (ITS) have established software system-enabled transportation infrastructure to the public using which traveller information service and hassle free transport have become the prime objective of the transport industry. At present, innovation in technology driven infrastructure planning in transportation management is highly demanded research prospect in the area of intelligent transportation systems and services. Research effort towards development of ITS with statistical and machine learning (ML) approaches applied in time series analysis for traffic forecasting is enormous. But, the outcome of such researches is still under refinement considering various practical difficulties. Hence, the objective of this survey is to present a detailed insight on evolution of traffic time series forecasting with broad classification of methods and detailed summary of their results. Finally, comprehensive review results are presented with directions to address the research challenges.

Keywords: short-term traffic prediction; traffic operations; non-parametric; parametric; machine learning technique; data mining.

DOI: 10.1504/IJLSM.2021.115068

International Journal of Logistics Systems and Management, 2021 Vol.39 No.1, pp.77 - 110

Received: 09 Apr 2019
Accepted: 03 Jun 2019
Published online: 18 May 2021 *

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