Roman holiday: Tourist heterogeneous preferences for walking path elements

This paper investigates tourists' preferences in Rome (Italy) with respect to their “last mile” transport walking experience. Stated preference experiments and discrete choice models are used to estimate tourists' preferences for alternative walking paths configurations, taking into account path characteristics and conditions (e.g. public services presence, interference with other modes of transport and thermal comfort). We estimate Multinomial logit (MNL) models on the base of 365 choice experiments. The most important attribute is cleanliness, followed by green spaces. The paper estimates both time elasticity and the “willingness to walk” with respect to different path attributes. Segmenting the sample and estimating different MNL models allows considering preference heterogeneity. Results show there are some differences within the sample. In fact, those who usually walk are willing to walk more than others given some improvements in the path. Besides, we also propose, on the base of the last walking path followed, a “tourist walking satisfaction index” linked to utility. This sheds light on tourists' walking experience in different zones of the study area. In particular, adding restrooms along the paths increases tourist walking satisfaction the most. The paper provides useful information for policy-makers on how to plan, design and manage walking networks.