Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Civil Engineering

Major Professor

Lee Han

Committee Members

Hamparsum Bozdogan, Arun Chatterjee, Tom Urbanik


Some basic concepts about traffic which are correct in theory may be misinterpreted in practice. Such misinterpretations may lead to a different direction from the ideal operation. This four-part dissertation is designated to examine fundamental concepts in traffic operation, and to validate the impact of randomness on control delays, cycle-length optimization, control types, and the peak-hour factor.

Control delays experienced by drivers is a critical performance measure on interrupted-flow traffic which involves movements at slower speeds and stops on intersection approaches, as vehicles move up in the queue or slow down upstream of an intersection. Since the basic term of control delay in a signalized intersection was originally from queueing analyses within a cycle, results from such models may be inaccurate due to the neglect of inter-cycle traffic variation. Besides, traffic is rare varying on the clock. Therefore, the peak-hour factor will be inaccurate to a certain degree if peak periods are placed on the clock.

All parts of this dissertation, except the first and the last, are independent papers for different professional journals, and are summarized as follows. Part II of this dissertation, “Impacts of Inter-Cycle Demand Fluctuations on Delay”, distinguishes between intra- and inter-cycle demand fluctuations and recognizes the potentially significant impact of delay underestimation when inter-cycle demand fluctuation is unaccounted for, as in all previous models. “Short or Long … which is Better? A Probabilistic Approach towards Cycle Length Optimization” in the third part of this dissertation proposes a framework to determine the optimal or near-optimal cycle length for signalized intersections based on the criterion with minimal control delays. The fourth part with title “A Trade-Off Framework for Determining the Best Control at an Intersection” in this dissertation uses the same criterion with minimal control delays to assist decision makers in the trade-off between signals and stop signs for an intersection. Part V of this dissertation, “Impacts of Misplaced Peak Intervals on PHFs”, argues about the significant difference among different ways to define the peak intervals, and distinguishes the differences between the “real” and “on the clock” peak-hour factors.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."