INCORPORATING TIME DEPENDENT LINK COSTS IN MULTI-STATE SUPERNETWORKS

The concept of a multi-state supernetwork is considered a promising approach to model multi-modal and multi-activity travel patterns. A derived feature of this approach is that a point-to-point path through the supernetwork represents a specific activity-travel pattern. However, current supernetworks are constructed in a static way in that the elapsed time and the disutility of a single link are fixed, which might cause inaccuracy in link costs in a dynamic environment. To make the supernetwork approach more powerful and practical, this study attempts to incorporate more fully the time dimension: travel links with both private vehicle and public transport are calculated on-the-fly, and parking costs are made duration-dependent. In addition, available time windows at activity locations are taken in account. The suggested improved supernetwork approach can better capture space-time constraints, and underlying activity-travel behavior. A label correcting algorithm is proposed to find the optimal activity-travel path.