A highway work zone design and traffic management decision system

The major objective of this research was to create an assessment system to support the highway work zone design process and selection of traffic management strategies to improve safety and optimize work zone traffic operation for work zone designers. More specifically, the evaluation system was design to provide quantified safety and mobility impacts for alternative work zone designs and traffic management strategies. Accordingly, this system consisted of the work zone crash analysis component and mobility performance analysis component. Work zone crash case analysis was conducted to study the work zone crash characteristics and investigate the trigger events leading to crashes. The results included the nature of work zone crashes, the main causations leading to crashes, and corresponding countermeasures. Additionally, linear regression analyses were conducted to examine the relationship between crash rate and traffic condition and work zone design related variables. It was found that the construction type (segment/bridge) was the significant factor for crash rates. Work zone mobility performance analysis provided the understandings for the work zone traffic flow characteristics to support the establishment of work zone simulation models. The field investigation was fulfilled to study the impacts of work zone configurations and several traffic management strategies on traffic flow operation and driver behavior with emphasis on work zone throughputs, delay, speed reduction, lane changing behavior, and detour decision. Based on the findings from field studies, simulation models were established to represent the work zone traffic flow patterns and driver behavior in VISSIM. These models were able to reflect speed reduction and lane changing behavior by customizing the features provided in VISSIM. To replicate the diversion patterns under congestion, a hybrid process integrating logistic regression analysis with micro-simulation was developed in COM interface supported by VISSIM. The algorithm can estimate diversion rates based on prevailing traffic condition and historical data and therefore dynamically determine the diversion in simulation. Furthermore, the conflict analysis based on simulation outputs provided an appropriate approach to assist the safety impact analysis given the current issues related to the work zone crash analysis including availability and quality of crash data and work zone configuration data.