Modelling public-transport users’ behaviour at connection point

Out-of-vehicle times were shown to be perceived as being more onerous than in-vehicle time by transit users when making transfers. The present study has two main objectives. The first objective is to determine the effects of uncertainty, in out-of-vehicle times during transfers, on transit users’ willingness to use transfer routes. The second objective is to determine the influence of out-of-vehicle facilities, offered by public-transport (PT) operators, on transit users’ perception of trip attributes related to transfers. A user preference survey was conducted at two major PT terminals in Auckland, New Zealand. The survey data was modelled using cumulative prospect theory and fuzzy logic. The results showed that for all trip attributes, except for comfort, transit users’ exhibited risk averse behaviour; users’ revealed greater preference for the transfer route with less uncertainty in the out-of-vehicle times. For comfort, transit users’ displayed risk-taking characteristics when the waiting time for a seat was less than 5min. Such findings suggest that increasing the consistency in out-of-vehicle times will increase attractiveness of transfer routes thus enabling a more efficient and integrated network of PT routes to result in enlargement of ridership. Policy makers and PT planners must focus on methods of reducing uncertainty in out-of-vehicle times during transfers. Analysis of transit users’ perception of trip attributes, given their current station, revealed statistical evidence of differences for two trip attributes, transfer waiting time and vehicle delay. Such findings indicate that transit users who are accustomed to better out-of-vehicle facilities have a lower tolerance for uncertainty in transfer waiting times and delay times. To the authors’ knowledge, this study provides for the first time in literature a comparison between the two cognitive models. The comparison revealed that CPT and fuzzy logic models are both capable of representing transit users’ decision making process. However, while CPT provides an indication of transit users’ preference for a transfer route, fuzzy logic is capable of providing a closer approximation of the proportion of transit users preferring a transfer route.