Effect of integration of bicyclists and pedestrians with transit in New Delhi

Document Type

Journal Article

Publication Date


Subject Area

place - asia, place - urban, mode - bus, mode - bike, mode - pedestrian, land use - impacts, technology - emissions, planning - surveys, planning - integration, ridership - behaviour


Random-taste Mixed Logit Model, Stated preference survey, Pedestrian transit access, Bicycle transit access


Traditionally, transportation mode shares in cities have been calculated separately for walking, bicycling and transit. However, it is well known that all transit trips have an access and egress component which is mostly executed through walking or bicycling. Hence, the choice of whether to choose transit for a particular trip depends as much on the walking or bicycling component of the trip as the transit component itself. A major source of inaccuracy in traditional mode share estimation models is the failure to identify this inherent dependence of transit trips on bicycling and walking. In New Delhi, where almost all access and egress trips for buses are made by bicycle/walking, the inaccuracies in mode share estimation could be more significant. This research aims to study behavioral effects of integrating bicycling and walking infrastructure with transit and provide predictions for outcomes of policy implementations modifying bicycle-to-transit or walk-to-transit environment in New Delhi. Four policy variables each are selected that affect pedestrian access to transit (sidewalk width, lighting, crossings and surrounding hygiene) and bicycle access to transit (bicycle lane, bicycle parking, bicycle sharing and on-board vehicle capacity) respectively. To gauge user behavior for hypothetical situations, stated preference survey data is collected through intercept surveys. 90 respondents were interviewed with upto 10 choice scenarios per individual with a total of 897 scenario responses (461 Pedestrian Infrastructure Scenarios +436 Bicycle Infrastructure Scenarios). Choice modelling is performed through a simple Multinomial Logit (MNL) model (in case there is no significant heterogeneity among individual preferences) and Random-Taste Mixed Logit model (to incorporate significant heterogeneity among various types of individual preferences). Modelling results showed that among pedestrian infrastructure, only presence of crossings could affect transit use and there is possibly significant heterogeneity in the population regarding use of sidewalks. Among bicycle infrastructure variables, presence of bicycle lanes and bicycle sharing is expected to positively impact transit use with no significant heterogeneity among the population. Finally, based on modelling results, three policy implementation scenarios are tested – presence of pedestrian crossings near all transit stops, introduction of bicycle lanes throughout the city and introduction of bicycle sharing system throughout the city. The scenario analysis shows possibility of considerable rise in transit mode share and GHG emission savings. This motivates further research to corroborate these findings with a larger sample, evaluation of viability of the ideas and possibly investigating implementation details.


Permission to publish the abstract has been given by Elsevier, copyright remains with them.


Transport Policy Home Page: