Chulsu Yang

Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Civil Engineering


Civil Engineering

First Advisor

James C Williams


To maximize the potential for successful managed lanes (MLs) operations, there are critical design issues that require special attention. Adequate ML access spacing and length of the ML access opening are required for the preservation of freeway capacity, level of service, and safety. The traffic weaving from an entrance ramp to the general purpose lanes across to a ML access point and vice versa are effectively modeled as a two-sided Type C weave. The 2000 Highway Capacity Manual (HCM) methodology was calibrated for the major weave without lane balance or merging (standard Type C weave) but not for the two-sided Type C weave. It can provide only the roughest of approximations when applied to a two-sided Type C weave. The process of developing guidelines for ML access spacing and the length of the ML access opening was achieved by a sensitivity analysis of weaving area performance measures (capacity and safety) with respect to these weaving distances. This was not feasible with field data because the replicated sampling of performance measures under various interesting flows and weaving distances are the keys for research success. With the limited availability of the 2000 HCM and field data, as discussed, a calibrated VISSIM simulation model was used as a standard to develop guidelines. In addition, this research developed analytical weaving models based on gap-acceptance theory for the development of ML weaving and access guidelines. Model calibration was the process of ensuring that the model duplicates realistic driving behaviors so that simulation outputs are credible. Data were collected along IH 635 (LBJ Freeway) in Dallas, which has four general purpose lanes plus a single ML in each direction. The data were collected at two sites on this freeway, allowing the second site to serve in the validation of the simulation model that was calibrated at the first site.In the development of simulation-based guidelines, the capacity and safety were improved with an increase in ML access spacing between 2000 and 4000 ft, but the length of the ML access opening between 1000 and 1500 ft had a minor (or no) impact on the capacity and safety of the ML weaving area. The ramp to ML volume (or the ML to ramp volume) was also a key factor for the capacity and safety of the weaving section. The minimum ML access spacing with various ramp to ML volumes or ML to ramp volumes was determined where the change in capacity and safety stabilized. Based on the analytical models, the critical factors affecting the distance required for completing the weave are the speed and flow of the weaving area. This indicates that the longest weaving distance is required under capacity conditions with no (or low) weaving volume since weaving vehicles significantly diminish the speed and capacity flow of weaving areas. In the development of analytical model-based guidelines, the minimum access spacing and ML opening length are determined at the target probability that 95% of weaving vehicles can complete their weave under the worst case traffic conditions.


Civil and Environmental Engineering | Civil Engineering | Engineering


Degree granted by The University of Texas at Arlington