Abstract
This research aims to investigate the difference of the travel behaviours of the elderly in different urban areas. On the basis of the results of cluster analysis, the nested logit (NL) model is adopted to investigate the travel decision of the elderly in central urban areas and non-central urban areas. Moreover, it is used to forecast the change of their travel behaviours by implementing the policies of age-friendly society. The results show that the elderly in central urban areas are inclined to decide travel modes in advanced of determining trip chains, while the trip chains are first determined in non-central urban areas. For the central urban areas, the policies mainly affect the choice of travel modes of the elderly, rather than the choice of trip chains. However, for the non-central urban area, the policies simultaneously influence the choice of travel mode and trip chains of the elderly. Based on the results of the forecast, some measures are proposed in this study to meet different travel requirements of the elderly in central urban areas and non-central urban areas.