Integrated Choice and Latent Variable model to evaluate the joint effectiveness of the introduction of a new light rail line and an informative measure

There is an ample documentation demonstrating how, during the last 30 years, the use of the car has reached an ever higher level of popularity for many different reasons, like personal convenience and comfort. Nevertheless, car use also generates various negative externalities, like air pollution, noise and personal injuries, whose abatement represents a challenge for every transportation planner. This has therefore led to a growing interest in implementing a variety of policies, both structural and behavioural, to attempt persuading people to reduce the use of their car, by shifting their travel behaviour towards a more sustainable one. However, one of the problems of behavioural interventions measures is the difficulty of properly quantifying their effectiveness, not only by using descriptive analysis, but also making use of predictive models that consider both objective and variables. The aim of this work is that of evaluating the effects of a combination of hard (introduction of a new light railway line) and soft measures (personalized travel plans, PTP) in the metropolitan area of Cagliari (Italy), using control groups and data collected before and after the implementation of the program. In particular, the data was gathered during the progress of the project called “Cittadella Mobility Styles”, which focused on the travel patterns of individuals going to (or away from) a university’s medical-scientific complex (“Cittadella Universitaria”). The final aim of the project was the promotion of the newly built light rail line as a sustainable travel alternative. The data of a total of 194 people was included in the sample analysed. In doing so, we estimate an Integrated Choice and Latent Variable (ICLV) model (Vij and Walker, 2016) to assess the effects of the characteristics of built environment, of demographic indicators and of some psycho-attitudinal variables, on the choice of switching to the new light railway. To the best of the authors’ knowledge, this should be the first example of ICLV model that considers all the aforementioned elements. To evaluate the effects of the soft measure, three dummy variables were considered. The first one refers to individuals who received a PTP; the second one is specific for the control group; the last one is instead used to indicate car users who did not belong in the previous categories. The modelling results show instead that the latent variables “aversion to public transport” and “attachment to the car” negatively influence the propensity to change travel behaviour. Another aspect that emerges is that the likelihood to change travel behavior strongly depends on the type of intervention, with people people who received the PTP more likely to change than people belonging to the control group. REFERENCES Vij, A., & Walker, J. L. (2016). How, when and why integrated choice and latent variable models are latently useful. Transportation Research Part B: Methodological, 90, 192-217.