Evaluation of Work Zone Design Strategies: Quantifying the Impact of Driver Behavior on Traffic Flow and Safety

This paper presents an algorithm that considers the influence of driver behavior on traffic flow and safety when drivers encounter and respond to changing roadway conditions found in work zones. The conditions modeled by the algorithm in this research effort are lane drops and the forced merges that are caused by the lane drop. Central to the algorithm is the integration of two concepts: driver familiarity and driver adaptability to the changing road conditions. The use of these concepts incorporates two notions: 1) drivers must manage interaction with both the roadway and other drivers; and 2) drivers exhibit varying preferences for early or late merges based on their willingness to respond to upcoming lane restrictions and their inclination to be passive or aggressive in forced merge situations. The algorithm assesses driver behavior at an agent-based level and was developed with the aid of a field study. The paper proposes that the algorithm be considered for use as a microscopic evaluation tool to quantify the impacts of different work zone strategies on traffic flow and safety.