01593606

Component

Carpooling and other forms of ride sharing are of growing significance in urban travel as environment- friendly and sustainable travel alternatives. With the widespread use of location sensing technology, spatial data is easily accessible and widely available. Massive volumes of spatial datasets provide opportunities to work on spatial data mining. Clustering vehicle trajectories enables identification of potential carpool routes. The objective of this paper is to detect common sub-paths in a road network through the analysis of trajectory data and to propose potential routes for carpool commuters. The clustering algorithm was applied to a large set of meso-level simulated trajectory data in the Chicago area. The identified sub-path clusters provide a basis for detecting likely carpool routes. The carpool participation sensitivity analysis and the simulation test for departure and arrival guaranteed routes demonstrated that carpool programs could contribute to congestion relief and travel time reduction.

This paper was sponsored by TRB committee AP020 Standing Committee on Emerging and Innovative Public Transport and Technologies.

01584066

16-7013

English

17p

2016

Transportation Research Board 95th Annual Meeting

Digital/other

Figures; References; Tables

Algorithms; Carpools; Cluster analysis; Data mining; Routing; Vehicle trajectories

Chicago (Illinois)

Data and Information Technology; Highways; Planning and Forecasting

Transportation Research Board Annual Meeting 2016 Paper #16-7013

TRIS, TRB, ATRI

Jan 12 2016 7:03PM