Off-campus UMass Amherst users: To download campus access theses, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Theses that have an embargo placed on them will not be available to anyone until the embargo expires.

Access Type

Open Access

Document Type


Degree Program

Civil Engineering

Degree Type

Master of Science in Civil Engineering (M.S.C.E.)

Year Degree Awarded


Month Degree Awarded



gap acceptance, critical gap, gap, driver behavior, traffic safety


Given that “driver error” is cited as a contributing factor in 93 percent of all crashes, understanding driver behavior is an essential element in mitigating the crash problem. Among the more dangerous roadway elements are unsignalized intersections where drivers’ gap acceptance behavior is strongly correlated to the operational and safety performance of the intersection. While a basic understanding of drivers’ gap acceptance behavior exists, several unanswered questions remain.

Previous work has attempted to address some of these questions, however to date the research has been somewhat limited in scope and scale due to the challenges of collecting high fidelity gap acceptance data in the field. This research initiative utilized software newly developed for this project to collect gap acceptance data on 2,767 drivers at 60 sites, totaling 10,419 driver decisions and 22,639 gaps in traffic. This large-scale data collection effort allowed many of these remaining questions to be answered with an improved degree of certainty.

This research initiative showed that naturalistic driver gap acceptance behavior can realistically be observed and accurately recorded in the field in real time using a newly developed software tool. This software tool and study methodology was validation using high fidelity video reduction techniques.

This research compared different methods of analyzing gap acceptance data, in particular determining critical gap, seeing that the method used significantly affects the results. Conclusions were draw about the merits of each of the ten analysis methods considered.

Through the analysis of the large data set collected, the research determined that there exist appreciable and identifiable differences in gap acceptance behavior across drivers under varied conditions. The greatest differences were seen in relationship to wait time and queue presence. If a driver has queued vehicles waiting behind them and/or has been waiting to turn for a long period of time, they will be more likely to accept a smaller gap in traffic.

Additionally, an analysis of gap acceptance as it relates to crash experience identified critical situations where a driver's gap acceptance behavior contributes to the occurrence of a crash. Characteristics of the driver such as gender and approximate age associated with specific crashes were examined. Teen drivers were identified as exhibiting aggressive gap acceptance behavior and were found to be overrepresented in gap acceptance related crashes. Ultimately, a better understanding of the driver and environmental factors that significantly contribute to increased crash risk will help guide the way to targeted design solutions.


First Advisor

Michael A. Knodler