Expansion of a small-scale car-sharing service: A multi-method framework for demand characterization and derivation of policy insights

Santhanakrishnan Narayanan, Constantinos Antoniou

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


When a car-sharing system is small, at its early stages, the modal split for the service is very low (e.g. <50 trips per day for the entire system). This results in a situation, where it is not possible to account for the demand for the service through traditional mode choice models. In addition, the business model of such a car-sharing service can be designed for serving special trips (e.g., trips to furniture stores). In several cities, this special group of trips are not adequately reflected in the existing demand models. Therefore, the objective of this research is to develop a data-driven methodological framework for finding the demand for such a system and characterise the users. To characterise the users, a multinomial logit model is employed. A linear regression model is used to estimate the average daily demand for the whole system, and a dirichlet regression to distribute the system level daily demand to individual stations. The estimation results of the multinomial logit model indicate the influence of socio-demographic characteristics (age, employment, income, education, household vehicle ownership), use of conventional modes (bicycle, PT and private car) and supply parameter (fleet size). Moreover, the results reveal that the service can improve transport equity. The linear regression model shows that the average daily demand increases with an increase in the number of stations. Furthermore, the demand varies according to days of the week and months (i.e., seasonal variation). Similarly, the shares of demand for individual stations is influenced by days of week. In addition, the shares are affected by the system level daily demand, i.e., as the demand increases, certain stations attract more users. Besides the aforementioned findings, operational and policy measures are suggested under the following categories: (i) Mobility as a Service (MaaS), (ii) Pricing strategy, (iii) Hybrid fleet, and (iv) Campaigns. The contributions from this study enable several cities to characterise the car-sharing service at early stages for expansion and also support the sustainable growth and integration of the service towards MaaS, leading to a better and efficient transportation system.

Original languageEnglish
Article number103438
JournalJournal of Transport Geography
StatePublished - Oct 2022


  • Car-sharing
  • Demand prediction
  • Mobility as a Service (MaaS)
  • Multi-method analysis
  • Shared mobility
  • Sustainable urban mobility


Dive into the research topics of 'Expansion of a small-scale car-sharing service: A multi-method framework for demand characterization and derivation of policy insights'. Together they form a unique fingerprint.

Cite this