Placement of guide signs on highway is an important task to field traffic engineers. It relates to many factors, such as traffic conditions, highway geometry and driver behavior, etc., and may vary from region to region. It is crucial to effectively determine the guide sign placement so that drivers on highway can take the desired exit timely and avoid any abrupt maneuvers. This paper presents models of utilizing lane change distance to derive the sign placement. The models are estimated using detailed vehicle trajectory data at a high resolution collected in naturalistic driving environment. The entire lane change process is divided into several sub-maneuvers (from one lane to an adjacent lane), while models for each sub-maneuver are presented separately. Instead of statically determining the sign placement, the proposed methodology takes various factors into consideration, including traffic flow characters, highway geometries, and vehicle interactions. Based on reasonable assumptions on drivers vision, advance distance of guide sign placement under prevailing traffic conditions is thereby estimated. The methodology provided in this paper can work as a supplemental reference to MUTCD for placing guide signs more efficiently.
Samenvatting