01627661

Component

Due to technological improvements in recent years, the available independent data sources for reporting travel time on freeway and arterial corridors has increased. Combining and fusing data from different sources and turning data into useful information is an important part of any integrated transportation system. A successful travel time fusion scheme improves real-time data coverage and accuracy, and is capable of reporting reliability by quantifying underlying uncertainties. This paper presents a multi-source real-time travel time fusion framework, based on Dampster-Shafer evidential theory (D-S theory). Data source credibility and real-time data variance extracted for each heterogeneous data sources are considered as time-dependent variables in the proposed model. The model is independent of underlying technology behind each sources and the number of input sources is not limited. An application of the fusion framework by using real-world data collected from three sources on a Maryland freeway corridor is presented and discussed. The results from case study show that the fusion scheme is successful in merging data along with reliability indicator, and its resolution can be adjusted with respect to the intended applications.

This paper was sponsored by TRB committee ABJ30 Standing Committee on Urban Transportation Data and Information Systems.

01618707

17-04974

English

19p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; References; Tables

Data fusion; Data quality; Real time information; Transportation corridors; Travel time

Maryland

Data and Information Technology; Highways

Transportation Research Board Annual Meeting 2017 Paper #17-04974

TRIS, TRB, ATRI

Dec 8 2016 11:54AM