Spatio-Temporal Analysis of On Demand Transit: A Case Study of Belleville, Canada

The rapid increase in the cyber-physical nature of transportation, availability of GPS data, mobile applications, and effective communication technologies have led to the emergence of On-Demand Transit (ODT) systems. In September 2018, the City of Belleville in Canada started an on-demand public transit pilot project, where the late-night fixed-route (RT 11) was substituted with the ODT providing a real-time ride-hailing service. We present an in-depth analysis of the spatio-temporal demand and supply, level of service, and origin and destination patterns of Belleville ODT users, based on the data collected from September 2018 till May 2019. The independent and combined effects of the demographic characteristics (population density, working-age, and median income) on the ODT trip production and attraction levels were studied using GIS and the K-means machine learning clustering algorithm. The results indicate that ODT trips demand is highest for 11:00 pm-11:45 pm during the weekdays and 8:00 pm-8:30 pm during the weekends. We expect this to be the result of users returning home from work or shopping. Results showed that 39% of the trips were found to have a waiting time of smaller than 15 minutes, while 28% of trips had a waiting time of 15-30 minutes. The dissemination areas with higher population density, lower median income, or higher working-age percentages tend to have higher ODT trip attraction levels, except for the dissemination areas that have highly attractive places like commercial areas.

[1]  Sokratis Basbas,et al.  A methodological framework for assessing the success of Demand Responsive Transport (DRT) services , 2016 .

[2]  M. Quddus,et al.  A survey of Demand Responsive Transport in Great Britain , 2012 .

[3]  Trupti M. Kodinariya,et al.  Review on determining number of Cluster in K-Means Clustering , 2013 .

[4]  Kari Watkins,et al.  Comparing Fixed-Route and Demand-Responsive Feeder Transit Systems in Real-World Settings , 2013 .

[5]  Xiugang Li,et al.  Feeder transit services: Choosing between fixed and demand responsive policy , 2010 .

[6]  Purnima Bholowalia,et al.  EBK-Means: A Clustering Technique based on Elbow Method and K-Means in WSN , 2014 .

[7]  Joseph Y. J. Chow,et al.  An agent-based day-to-day adjustment process for modeling ‘Mobility as a Service’ with a two-sided flexible transport market , 2017 .

[8]  J. Nelson,et al.  DEMAND RESPONSIVE TRANSPORT: TOWARDS THE EMERGENCE OF A NEW MARKET SEGMENT , 2004 .

[9]  Chaita Jani,et al.  Implementing & Improvisation of K-means Clustering Algorithm , 2016 .

[10]  Jari Saramäki,et al.  Where did Kutsuplus drive us? Ex post evaluation of on-demand micro-transit pilot in the Helsinki capital region , 2019, Research in Transportation Business & Management.

[11]  Marcus P. Enoch,et al.  Multilevel modelling of Demand Responsive Transport (DRT) trips in Greater Manchester based on area-wide socio-economic data , 2014 .

[12]  A. Khattak,et al.  TRAVELER RESPONSE TO INNOVATIVE PERSONALIZED DEMAND-RESPONSIVE TRANSIT IN THE SAN FRANCISCO BAY AREA , 2003 .

[13]  Philip S. Yu,et al.  Top 10 algorithms in data mining , 2007, Knowledge and Information Systems.

[14]  Sukavanan Nanjundan,et al.  Identifying the number of clusters for K-Means: A hypersphere density based approach , 2019, ArXiv.

[15]  Lauri Häme,et al.  Demand-Responsive Transport: Models and Algorithms , 2013 .

[16]  P. Mokhtarian,et al.  What drives the use of ridehailing in California? Ordered probit models of the usage frequency of Uber and Lyft , 2019, Transportation Research Part C: Emerging Technologies.

[17]  Zeina Wafa,et al.  Assessing the Impact of App-Based Ride Share Systems in an Urban Context: Findings from Austin , 2018, Transportation Research Record: Journal of the Transportation Research Board.

[18]  David Koffman,et al.  Operational Experiences with Flexible Transit Services , 2004 .

[19]  Steven Farber,et al.  The Who, Why, and When of Uber and other Ride-hailing Trips: An Examination of a Large Sample Household Travel Survey , 2019, Transportation Research Part A: Policy and Practice.

[20]  Maria Grazia Speranza,et al.  A simulation study of an on-demand transportation system , 2018, Int. Trans. Oper. Res..

[21]  José Antonio Lozano,et al.  An efficient K -means clustering algorithm for massive data , 2018, ArXiv.