Urban road networks generally consist of a large number of intersections. Drivers perform aturn maneuver at each intersection they approach as occasions demand. Delays which driversexperience at intersections also may vary with traffic volume. Therefore, a shortest pathalgorithm for urban road networks should provide drivers with travel time informationconsidering turn penalty varying with traffic volume. The purpose of this paper is to develop ashortest path algorithm which considers turn penalty in time-dependent networks. Kaufman?salgorithm, which is one of well-known shortest path algorithms, is modified in this paper toconsider turn penalty. Gangnam-gu which is one of the most crowded areas in Seoul isselected as our study site. Link travel times per minute are calculated by PARAMICS, themicroscopic traffic simulation software. A comparison is performed between the minimumtravel costs calculated by Dijkstra?s algorithm, the original Kaufman?s algorithm and themodified Kaufman?s algorithm. The result shows that total travel time searched by thealgorithm in this study is different from them by the other existing algorithms and thus theshortest path information considering turn penalty should be provided This research and the result will not only maximize drivers? benefit by offering reality-based travel information, butalso by offering an alternative route.
展开▼
