Skip to main content
placeholder image

Analysing the merit of latent variables over traditional objective attributes for traveller mode choice using RPL model

Conference Paper


Download full-text (Open Access)

Abstract


  • In real life, the attributes that influence individual choice may be complex. The traditional objective attributes can be incorporated easily into choice models. However, there are also latent preference heterogeneities that are of ten overlooked by the traditional thinkers of transport planners. This paper deals with this issue firstly by testing the adequacy of objective attributes representing latent variables (LVs). It then quantifies the effect of LVs over objective attributes on traveller mode choice using the random parameter logit (RPL) model. Understanding these attributes is essential if transport agencies are to understand traveller behaviour when determining effective transport policies. This paper emphasises travellers’ LVs along with objective attributes during the mode choice process as a method by which the

    utility of the traveller can be maximised. Thus the issue of utility function is raised and evaluated using a discrete choice experiment, i.e. RPL model. An empirical study was carried out in the context of traveller behaviour in the Sydney Statistical Division (SSD). We consider six LVs and thirteen objective attributes to analyse the importance/merits of LVs over objective attributes in traveller mode choice. The results show that indicators of LVs and traveller choice attributes are found to be significant, while objective attributes show a very minimal (0% to 10% on average) capacity to reflect LVs in traveller choice processes. LVs are found to be more influential than objective attributes on the mode choice made by travellers and our results also show that hybrid RPL is superior to traditional RPL models that ignore the effect of LVs. Our results support the contention that latent factors are important in traveller mode choice in ways that are relevant to transportation planners and policy-makers. Although possibly not directly susceptible to policy intervention, a better understanding of

    these relationships is useful for decision makers and transportation planners when designing and developing sustainable transportation policies or projects for the city dwellers.

Publication Date


  • 2013

Citation


  • Anwar, A. H. M. Mehbub., Tieu, K., Gibson, P., Win, K. Than. & Berryman, M. J. (2013). Analysing the merit of latent variables over traditional objective attributes for traveller mode choice using RPL model. International Choice Modelling Conference (pp. 1-23). United Kingdom: ICMC.

Ro Full-text Url


  • http://ro.uow.edu.au/cgi/viewcontent.cgi?article=3556&context=eispapers

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers/2547

Start Page


  • 1

End Page


  • 23

Place Of Publication


  • United Kingdom

Abstract


  • In real life, the attributes that influence individual choice may be complex. The traditional objective attributes can be incorporated easily into choice models. However, there are also latent preference heterogeneities that are of ten overlooked by the traditional thinkers of transport planners. This paper deals with this issue firstly by testing the adequacy of objective attributes representing latent variables (LVs). It then quantifies the effect of LVs over objective attributes on traveller mode choice using the random parameter logit (RPL) model. Understanding these attributes is essential if transport agencies are to understand traveller behaviour when determining effective transport policies. This paper emphasises travellers’ LVs along with objective attributes during the mode choice process as a method by which the

    utility of the traveller can be maximised. Thus the issue of utility function is raised and evaluated using a discrete choice experiment, i.e. RPL model. An empirical study was carried out in the context of traveller behaviour in the Sydney Statistical Division (SSD). We consider six LVs and thirteen objective attributes to analyse the importance/merits of LVs over objective attributes in traveller mode choice. The results show that indicators of LVs and traveller choice attributes are found to be significant, while objective attributes show a very minimal (0% to 10% on average) capacity to reflect LVs in traveller choice processes. LVs are found to be more influential than objective attributes on the mode choice made by travellers and our results also show that hybrid RPL is superior to traditional RPL models that ignore the effect of LVs. Our results support the contention that latent factors are important in traveller mode choice in ways that are relevant to transportation planners and policy-makers. Although possibly not directly susceptible to policy intervention, a better understanding of

    these relationships is useful for decision makers and transportation planners when designing and developing sustainable transportation policies or projects for the city dwellers.

Publication Date


  • 2013

Citation


  • Anwar, A. H. M. Mehbub., Tieu, K., Gibson, P., Win, K. Than. & Berryman, M. J. (2013). Analysing the merit of latent variables over traditional objective attributes for traveller mode choice using RPL model. International Choice Modelling Conference (pp. 1-23). United Kingdom: ICMC.

Ro Full-text Url


  • http://ro.uow.edu.au/cgi/viewcontent.cgi?article=3556&context=eispapers

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers/2547

Start Page


  • 1

End Page


  • 23

Place Of Publication


  • United Kingdom