Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Civil Engineering


Civil Engineering

First Advisor

C Williams C James Williams

Second Advisor

Behruz Paschai


Systematic, well-designed research provides the most effective approach to the solution of many problems facing highway administrators and engineers. Traffic congestion on freeway systems is one significant concern in urban areas throughout the U.S.A. In this era, building new freeways to reduce congestion is less feasible due to the high capital and social costs. Thus, the effective management and operation of existing freeway facilities has become a preferred approach to reduce traffic congestion. A weaving section is a common design on major highway facilities that always has been an interest to researchers. Weaving areas are characterized by frequent lane changes, which significantly reduce the capacity of the freeway system. The HCM defined weaving capacity as “any combination of flows that causes the density to reach the LOS E/F boundary condition of 43 pc/m/ln for freeways” based on configuration, number of lanes in the weaving section, free-flow speed, length of the weave, and volume ratio (VR). Weaving sections are often problematic because of the increase in lane changing. These are common design elements on freeway facilities between an on ramp and off ramp with an auxiliary lane. They are located between merge and diverge points, near ramps where a lane is added or dropped, and at multilane ramps. A weaving section is a freeway segment in which traffic flows cross each other without traffic control (Minderhoud et al. 2003). Traffic demands exceeding segment capacity at weaving areas cause congestion, which affects the operation of the entire freeway section. Traffic operation problems often exist at weaving areas even when traffic demands are less than capacity and may be experienced at lower traffic flows because of the complexity of vehicle interactions, that is, increased lane changing, resulting in a degradation in level of service (LOS) and potential safety problems (Skabardonis et al. 2012). A significant amount of research has been done to estimate quality of service and capacity in weaving sections. However, little has been done to address multiple weaves. A multiple weaving area is one where two or more weaving areas overlap. No satisfactory means of estimating capacity has been found. Current procedures make assumptions about where weaving occurs in the individual weaving segments. A multiple weaving area is found when “a series of closely spaced merge and diverge areas create overlapping weaving movements (between different merge-diverge points).” (HCM 2016) This work examines capacity and quality of service conditions for a specific example of a multiple weave. Capacity is evaluated through micro-simulation by gradually raising flows for a range of geometric and fraction of weaving-traffic conditions. Models are developed to express capacity in terms of lane configuration, flow ratios, traffic mix (heavy traffic percentages), and overall flow rate. This study will be limited to cases where there are two overlapping weaving movements created by two entry ramps following by an exit ramp. The study will develop relations for capacity and service volume for a range of geometric, flow conditions, and traffic mix. Geometric conditions will include number of lanes on the main lane entry and each entrance ramp into and exit ramp from the weaving area (n), and distances (Ls) between ramps. Flow conditions will include a range of flows from each of the entry and exit roadways (V) in the multiple weaving area. The traffic mix represents the fraction of heavy vehicles in the traffic stream (PHV). In order to estimate the capacity of the weaving area, repeated simulations are performed under varying factors, number of lanes, flow range, heavy vehicle percentage, and routing. VISSIM outputs include link evaluation (average values for the link), and data collection (point measurement) files are extracted. These output files allow us to extract parameters, such as space-mean speed for links, to identify the point at which speed start to drop. Also, entry and exit volumes to see if they are the same as the summation of input volume that we specified in the run. The point that speed starts dropping significantly, or the summation of exit or entrance volume is much lower than what we entered in the system, suggests that we have reached the capacity point. VISSIM simulation runs are done using COM programming. This VISSIM scripting tool will enable the user to automate the analysis process. Using COM programming, this research has been able to performs many more simulation iterations than otherwise would have been possible in the same time period. Not only for the factors but also for the driving behavior parameters (following, lane change, and car following). To estimate capacity, we will consider all possible combinations of car following parameters to see what gives us the maximum capacity. At the same time, we will adjust different variables/factors we have introduced (number of lanes, flow, etc.) to see which one will give us a higher throughput, which is the capacity. This work’s result will be presented to DOTs and MPOs as a guidebook. The file would be highly useful and money-saving for these agencies as they prefer to obtain higher capacity by managing existing freeways rather than buying rights of ways


Multiple weaving areas, Capacity, VISSIM


Civil and Environmental Engineering | Civil Engineering | Engineering


Degree granted by The University of Texas at Arlington