01479246

Component

Extracting structured knowledge from large datasets is a challenge that transportation planning faces more and more, particularly in the context of assisting policy decision-making. This paper takes advantage of the availability of 'BIG' operational archived data to scrutinize weekly travel variability of paratransit users and extract representative patterns of weekly travel behaviour. To that end, data mining techniques are used. The study is based on a one-year dataset which represents 1,393,291 trips made by 10,182 users. Empirical results show that the week structure regulates the activity rhythms of the paratransit system with more than 92% of successive daily trip chains performed inside a seven-day period. This study also confirms that weekly rhythms of paratransit are different from the weekly rhythms of the general population. In the future, paratransit planning should move towards a more data- and user-oriented paradigm if it intends to achieve an allocation of operational resources more commensurate with its ridership.

This paper was sponsored by TRB committee AP060 Paratransit.

01470560

13-3790

English

19p

2013

Transportation Research Board 92nd Annual Meeting

Digital/other

Figures; References; Tables

Data mining; Decision making; Paratransit services; Policy making; Ridership; Transportation planning; Travel behavior; Trip chaining; Weekly

Planning and Forecasting; Public Transportation; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2013 Paper #13-3790

TRIS, TRB, ATRI

Feb 5 2013 12:44PM