Understanding the demand for carsharing: lessons from Italian case studies

The aim of this paper is to estimate the potential demand for carsharing, to this aim a model which calculates the total generalized cost for a given mobility pattern and transport mode mix is developed. The model considers: a) that a person sometimes travels with friends and family, and therefore shares the travel expenses and/or satisfies several travel needs, and b) that uses in given time period more than one mode of transport. The parameters of the model are derived by detailed, face-to-face, computer-assisted interviews. A limited number of interviews have been so far completed. However, they hint to some very interesting empirical evidence. It is found that car ownership is currently very high in the Italian families and that the car is used extensively both for work\study and, especially, for other-than-work\study trip purposes. Offering a carsharing service (CS) would enhance the mode choice and could, in some cases, lower the total mobility costs. The respondents assign quite a large value to the pleasure of owning a car, much more so than the pleasure of being carsharing users, both for workers and for students. Consequently, the respondents would dislike not owing a private car, while having the choice between the private and the carsharing car is preferred especially by the students. The mobility cost indicators reflect, but not perfectly, the preference-based choices of the sample. Three individual case studies are further analyzed. They have been defined as: a low, a medium and a high mobility case study. The low mobility case study shows that these persons would largely benefit from the existence of a CS service, they would use it occasionally and they would probably be willing to forgo the private car. The medium mobility case study shows that the variables parking time, access time and CS fare can easily switch the balance between convenience and inconvenience of using CS. The higher mobility case study in a small town setting demonstrates that in such circumstances the prospects for a viable CS service are rather bleak.