Theses and Dissertations


Hohyun Lee

Issuing Body

Mississippi State University


Strawderman, Lesley

Committee Member

Giesen, James M.

Committee Member

Usher, John M.

Committee Member

Jin, Mingzhou

Committee Member

Yuan, Changhe

Date of Degree


Document Type

Dissertation - Open Access


Industrial Engineering

Degree Name

Doctor of Philosophy


James Worth Bagley College of Engineering


Department of Industrial and Systems Engineering


Walking is the oldest and foremost mode of transportation through history and the prevalence of walking has increased. Effective pedestrian model is crucial to evaluate pedestrian facility service level and to enhance pedestrian safety, performance, and satisfaction. The objectives of this study were to: (1) validate the efficacy of utilizing queueing network model, which predicts cognitive information processing time and task performance; (2) develop a generalized queueing network based cognitive information processing model that can be utilized and applied to construct pedestrian cognitive structure and estimate the reaction time with the first moment of service time distribution; (3) investigate pedestrian behavior through naturalistic and experimental observations to analyze the effects of environment settings and psychological factors in pedestrians; and (4) develop pedestrian level of service (LOS) metrics that are quick and practical to identify improvement points in pedestrian facility design. Two empirical and two analytical studies were conducted to address the research objectives. The first study investigated the efficacy of utilizing queueing network in modeling and predicting the cognitive information processing time. Motion capture system was utilized to collect detailed pedestrian movement. The predicted reaction time using queueing network was compared with the results from the empirical study to validate the performance of the model. No significant difference between model and empirical results was found with respect to mean reaction time. The second study endeavored to develop a generalized queueing network system so the task can be modeled with the approximated queueing network and its first moment of any service time distribution. There was no significant difference between empirical study results and the proposed model with respect to mean reaction time. Third study investigated methods to quantify pedestrian traffic behavior, and analyze physical and cognitive behavior from the real-world observation and field experiment. Footage from indoor and outdoor corridor was used to quantify pedestrian behavior. Effects of environmental setting and/or psychological factor on travel performance were tested. Finally, adhoc and tailor-made LOS metrics were presented for simple realistic service level assessments. The proposed methodologies were composed of space revision LOS, delay-based LOS, preferred walking speed-based LOS, and ‘blocking probability’.



cognitive information processes||M/G/c/c queues||pedestrian behavior||level of service