Effects of residential move on interaction frequency with social network

Size and composition of social networks have recently been identified in the transportation literature as important triggers of travel demand in general and travel generation in particular. With the field of activity-based modeling of travel demand moving from static to dynamic models, this implies that changes in social networks should be analyzed and linked to such dynamic activity-based models. Changes in social networks in turn are understood to be related by life trajectory decisions. Using a Dutch Kinship Panel dataset, the aim of this paper is to contribute to this emerging topic. Bayesian belief networks are used to analyze the direct and indirect relationships of one life trajectory decision (residential move) on changes in kinship networks. This is a necessary building block in developing comprehensive dynamic activity-based models of travel demand. Results support the feasibility of the suggested approach