Defining the role of mobility as a service – A discrete choice framework incorporating missing data

Mobility as a Service (MaaS) is an emerging concept that integrates various sustainable and shared modes using a digital interface. Currently, it is promoted as an option that encourages shift from private vehicles. Its preference among private vehicle non-users is unexplored. The current study models the willingness of individuals to choose a Maas package consisting of Mass Rapid Transit System (MRTS), bus and shared bicycle options in the city of Chandigarh, India using an Integrated Choice and Latent Variable (ICLV) framework. It addresses the issue of missing income, in the data collected, by assuming the income variable to be latent in the ICLV model. The structural and measurement equations of the latent income variable are specified using novel error formulations having either normal or lognormal distributions. Among the models estimated, multiplicative models with lognormal error distribution in the structural and measurement equations were observed to result in a better fit than the conventional model with normal error distributions in both structural and measurement equations. Across models, income and environmental consciousness were found to significantly increase the preference for MaaS. Further, our results highlighted a need for improving the quality of existing shared modes and public transport that are included in the MaaS subscription. The preference for MaaS among low-income respondents, and females, who do not have a driving license, indicated that MaaS can be considered as an option to address the issue of transportation inequity, especially for people having low chances of owning a private vehicle.