Binary Logistic Model for Estimation of Mode Shift into Delhi Metro

Abstract: This paper aims to study the public transport mode choice behaviour of commuters in Delhi so that appropriate strategiesto incentivize the use of public transport can be developed. We examine the efficacy of a multivariate statistical modelling approachto predict the probability of non-metro commuters to shift to the Delhi metro. We also analyse the reasons for this shift from privatemotor vehicles (PMVs) and buses. Data is collected through a survey of the metro commuters over various metro lines. A binomiallogistic regression model is formulated to predict whether existing metro users have shifted from buses or are new additions to publictransport shifting from PMVs. The model is validated well through several methods. The model analysis reveals that 57% of themetro users have shifted from buses and 28.8% from PMVs. The shift is more amongst females than males