Back close

Prediction of Traffic Volume by Mining Traffic Sequences Using Travel Time based Prefix span

Publication Type : Journal Article

Publisher : IET Intelligent Transport Systems

Source : IET Intelligent Transport Systems, 2019

Url : https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/iet-its.2018.5165

Campus : Chennai

School : School of Engineering

Department : Electronics and Communication

Year : 2019

Abstract : Traffic pattern analysis is an active and essential part of transportation research. When traffic condition is adverse and unprecedented, traffic sequences are useful in the analysis of traffic behaviour. The sequence through which traffic congestion has arisen can be predicted using sequence rules from the generated traffic sequence. This work aims at mining traffic sequence pattern and prediction of traffic volume based on traffic sequence rules. To mine peak hour traffic sequences in order to make better travel decision, travel time based PrefixSpan (TT-PrefixSpan) algorithm is proposed to analyse traffic flow on highways. As a result, the prediction of traffic volume is effected by the generated traffic sequences. Such analysis would pave the way for devising data driven computational methods in reducing traffic congestion. Real-time traffic volume data for 53 weeks is collected at a centralised toll system comprising toll collections centres at three different sites. To show the significance of this problem-solving approach, TT-PrefixSpan is experimented on three different sites. The extraction of a frequent traffic sequence pattern is reported with experimental analysis. The evaluation of different traffic condition present at each site has shown promising results. Towards the end, a summary of results is presented with directions for future research.

Cite this Research Publication : Prediction of Traffic Volume by Mining Traffic Sequences Using Travel Time based Prefix span, IET Intelligent Transport Systems, , Print ISSN 1751-956X, Online ISSN 1751-9578, Volume13, Issue7, pp.1199 –1210, Available online from 1st April 2019 and Print ISSN from 1st July 2019, England, Impact factor 2.05.

Admissions Apply Now