An in-depth study of urban freeway onramp merge behavior and operational characteristics using videotaping and computer-aided image processing and data reduction methods

Observational data were collected at urban freeway onramps along I-35 within the Kansas City metropolitan area by videotaping, and processed using computer-aided image processing and data reduction methods. Based on different freeway merging behaviors, three types of merge were categorized. These are: free merge (FM), challenged merge (CM), and platoon merge (PM). Data analysis results showed that: (a) The freeway merge process is a two-way influence phenomenon: not only influences from the freeway flow to the ramp merging flow, but also influences from the ramp merging flow to the freeway flow. The latter was deemed as invasive-influences in this research, because the freeway vehicles' priority is violated and the freeway flow interrupted. The significance of the invasive-influences was introduced as more straightforward level of service (LOS) indicators that can be used to evaluate the operational efficiency of freeway onramps. (b) The PM volume was observed to be most indicative of merge impacts of ramp vehicles on freeway traffic, especially on freeway right-hand Lane 1 traffic. Therefore, invasive platoon merge was identified as the potential cause of freeway delay or congestion that often develops in the vicinity of a freeway onramp. (c) The CM vehicles always require traveling a longer distance before merge. The necessary merge length (distance between the merge point and the ramp nose) of a majority of CM vehicles is critical for determining the onramp merge lane length. (d) When the volume of freeway Lanes 1 and 2 exceeds roughly 3,000pcph, a ramp driver will rarely get a chance to make a free merge (in this case, the percentage of FM volume approaches zero). Consequentially, the 85th percentile merge length of all ramp vehicles was found to be related to the PM volume, the CM volume, and the freeway Lane 1 average speed.; Finally, the operational and design implications of these findings were discussed. As a result, a CM-based method and procedure was proposed for the adequate design of urban freeway onramps with moderate to high merging volumes. A numerical example was also presented, showing how to determine the minimum merge lane length for urban freeway onramps.