You are here

Evaluating Citywide Bus Service Reliability Using Noisy GPS Data


Shen Wang, Brian MacNamee

Publication Type: 
Refereed Conference Meeting Proceeding
Abstract—An increasing number of people use smartphone applications to plan their trips. Unfortunately, for various reasons, bus trips suggested by such applications are not as reliable as other trip types (e.g. by car, on foot, or by bicycle), which can result in excessive waiting time, or even the need to revise a planned trip. Traditional punctuality-based bus service reliability metrics do not capture route deviations, which are especially frequent in rapid changing urban environments due to rapidly changing road conditions caused by traffic congestion, road maintenance, etc. The prevalence of GPS data allows buses to be tracked and route deviations to be captured. We use such data to propose and calculate a novel reliability score for bus trips. This score is a linear weighted combination of distance, time, and speed deviations from an expected, predefined bus trip. GPS trajectory data is large and noisy which makes it challenging to process. This paper also presents an efficient framework that can de-noise and semantically split raw GPS data by pre-defined bus trips in citywide. Finally, the paper presents a comparative case study that applies the proposed reliability score to publicly available open bus data from Rio de Janeiro in Brazil and Dublin in Ireland.
Conference Name: 
IEEE International Smart Cities Conference (ISC2) 2017
Digital Object Identifer (DOI): 
Publication Date: 
National University of Ireland, Dublin (UCD)
Open access repository: