Date of Award

8-2016

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Civil Engineering

Committee Member

Dr. Jennifer H. Ogle, Committee Chair Dr.

Committee Member

Bradley J. Putnam

Committee Member

Dr. Wayne Sarasua

Abstract

More than 3,500 deaths and thousands of injuries occur every day on roads all over the world. The International Road Assessment Program (iRAP) states, “Currently 90% of the world’s 1.25 million road fatalities per annum are in low and middle income countries, and by 2020 the number of road fatalities in these countries is expected to grow by 50%.” The compound problem in developing nations stems from roads which are rapidly constructed without much regard to proper design or safety, a lack of attention to vulnerable road users, and the absence of road safety culture (i.e., safe behaviors, vehicle safety regulations, road safety policy, road safety assessment, and enforcement). In Haiti, the road safety problem is exacerbated by the lack of data related to roadway crashes and the resulting fatalities and injuries. In numerous international road safety reports by organizations such as the World Health Organization, World Bank, and others, Haiti is one of the few Latin American Countries (LACs) that is not represented with national road safety and fatality statistics due to the limited availability of safety data. Some of the data issues can be attributed to the 2010 earthquake that destroyed much of the capital city of Port-au-Prince, Haiti. The Inter-American Development Bank released a project statement in 2010 that contained a glimpse at road safety in Haiti. According to the documentation, the Office d’Assurance Vehicules Contre Tiers (OAVCT) indicated only 108 fatalities in all of Haiti. This limited data is likely a function of limited insurance coverage among motorists in Haiti. Conversely, an NGO, INGO, operating a medical facility in Haiti noted 52 fatalities and 376 injuries in only 55 days on a 20 km section of National Highway 2 between Léogâ ne and Gran d Goâ ve. Given that there are a total of 6045 km of National Highways in Haiti; a fatality rate similar to this for the rest of the country would indicate roughly 15700 deaths due to roadway crashes. Some number between 108 and 15700 is likely the true answer. INGO also indicated that in 32 cases where occupants survived amputations were necessary, leaving crash victims with lifelong disabilities. Meta-analysis of similar reports indicate that approximately 50% of the trauma cases seen in the Haiti hospitals are related to transport crashes. These issues will only get worse with OAVCT reporting growth in motorization of roughly 10% per year. To combat these road safety issues, this research is undertaken in conjunction with the International Road Assessment Program (iRAP), whose goal is “a world free of high-risk roads.” While the long-term goal of this project is to create an iRAP presence across Haiti by assessing road and safety conditions through road analysis programs, this thesis covers the initial setup, training, implementation, and coding evaluation. Road video data and GPS data were collected along National Route 3, from Port-au-Prince to Cange in addition to gathering speed data in Domond and video data of pedestrian traffic in Cange. This data, gathered in areas that were deemed to be high-risk, were used to provide an idea concerning t traffic in the area. The data was processed and analyzed using FPZ, an iRAP road analysis program developed by University of Zagreb in Croatia, where the videos were processed along the route and road centerlines were created and segmented with respect to the video files for each segment. Data analysis was followed by iRAP road coding, for all the road sections based on 52 different road attributes. QA check of the coded data generated numerous errors prompting the need to develop a Haiti-specific iRAP coding manual to train the raters involved in this project. Inter-rater reliability tests were carried along with Cohen’s Kappa statistic to assess the agreement among the raters and accuracy with respect to iRAP coding standards. These tests and the assessment of reliability helped the raters to understand the coding process better, and get a good grasp of roadside attributes present along the project route in Haiti. This lays a good foundation for future research and further assessment of the route, which involves generating star-ratings of the road sections upon successful road coding. This report and its implementation greatly assisted the team involved with this project in learning the intricacies of the correct iRAP coding techniques, which has laid the foundation to go further with the ultimate goal of obtaining star ratings of the sections, indicating the high, medium and low risk road sections.

Share

COinS