출근시간대 차로별 속도특성을 이용한 도시고속도로 기본구간의 지체추정에 관한 연구: Michigan주 I-696을 대상으로

Abstract

Generally urban freeway means the intra-city freeway with 4 lanes or more, which plays a key role in the urban transportation system, carries the tremendous automobiles, and keeps the free flow speed except for the rush hours. Urban freeway (I-696) under this study is a divided ground-level freeway having 3 to 4 lanes in each direction from Novi in the west to St. Clair Shores in the east in the State of Michigan, and is also suffering from severe traffic congestion during the morning rush hours. So it was absolutely needed to study the appropriate travel characteristics on urban freeway basic segment congested during the rush hours.

The purpose in this study was to investigate roadway and traffic characteristic data on the 8-lane urban freeway basic segments congested during the rush hours (06:30~07:30), analyze the correlation between the lane-based traffic characteristics in the urban freeway basic segments congested, and finally suggest the optimal delay model for predicting travel delay on the urban freeway basic segments using the lane-based speed characteristics.

From the traffic characteristic analyses, and development and validation of delay model in urban freeway, the following conclusions were drawn

ⅰ) Urban freeway traffic management system(UFTMS) using the lane-based traffic characteristics was thought to need for increasing traffic safety, reducing traffic congestion, and maximizing lane utilization because there was a distinct difference in the lane-based traffic characteristics.

ⅱ) Urban freeway delay prediction technique using the lane-based traffic characteristics was thought to need for speed control on the urban freeway because there was a negative correlation between the lane-based speeds and travel delay in freeway, especially a high correlation between the 2nd lane speed and travel delay in urban freeway.

ⅲ) Cubic model and exponential model proved to fit for predicting travel delay with the 2nd and 3rd lanes, and the 4th lane in urban freeway basic segment, respectively, but especially cubic model proved to fit for best predicting travel delay in urban freeway with the higher explanatory power of the determination coefficients() of 0.941.

ⅳ）Cubic model in urban freeway especially proved a high correlation between observed delay and expected delay with the high correlation coefficients() of about 0.8124 to 0.9816, and also proved a high explanatory and valid delay predictive model with the low RMSE results of about 1.4 sec to 4.1 sec between the delay models using the lane-based speed characteristics.