Type of Document	Thesis
Author	Saravade, Snehal
Author's Email Address	saravsn@eng.fsu.edu
URN	etd-01042005-233714
Title	A Study of Fatal Rollover Crashes in the State of Florida
Degree	Master of Science
Department	Civil and Environmental Engineering, Department of
Advisory Committee	Advisor Name Title Lisa K. Spainhour Committee Chair Makola Abdullah Committee Member Primus V. Mtenga Committee Member
Keywords	Rollover Overepresentation Factor Traffic Crashes SUV Non-SUV
Date of Defense	2004-11-10
Availability	unrestricted
Abstract ABSTRACT The highway fatality rate in Florida remains unacceptably high. The Florida Department of Transportation’s Safety Office proposed conducting research on the fatal crashes on state roadways of Florida. The data set contained only fatal Commercial Motor Vehicles (CMV) crashes for the years 1998 and 1999; and all fatal crashes for the year 2000. This research investigates the causative factors for the rollover crashes with a special focus on the rollover crashes involving sports utility vehicles (SUV’s). A literature review found that despite crash test results that indicate a tendency to rollover in SUV’s, little research has focused on SUV behavior in actual crash conditions. This may be because many crash reports do not differentiate SUV’s as a distinct vehicle type. The primary purpose of this study was to determine effect of vehicle choice and other crash characteristics on fatal crash involvement and survivability between SUV’s and non-SUV’s. The non-SUV’s involve the vehicle types such as automobiles, passenger vans, pickup trucks and CMV’s. The research investigates the driver, vehicle, environment and roadway factors related to the causation of crashes and fatalities in SUV’s and non-SUV’s. The primary approach used in this study is the case-by-case review of rollover crashes. Every case was studied in detail to find the contributing factors that led to crash, such as driver age, distraction, inattention, speeding, vehicle defects and so on. Statistical inference tools have been used, such as hypothesis testing, confidence intervals and a simplified method of frequency distribution called overrepresentation factor. The overrepresentation factor is the ratio of percentage of the subset to the complement of the subset. Also, several summary and cross tab analyses were performed to investigate the correlation between different variables in the fatal rollover crashes. It was concluded with 95 percent confidence that SUV’s are three times likely to rollover than the non-SUV’s. The rollover fatality rate of SUV’s was significantly higher than the other type of crashes in the SUVs. But it was observed that, if involved in a rollover, SUV’s have lower fatality rates than non-SUV. In examining the factors contributing to rollovers SUV’s are highly overrepresented in tire blowout and tire tread separation crashes. Even though there are more males driving SUV’s, females are overrepresented in rolling over SUV’s. It was also found that younger and older drivers are rolling over SUV’s more often than the middle-aged drivers. There was significantly less alcohol/drug abuse on part of SUV’s drivers involved in crash than the non-SUV’s drivers, which is indicative of the fact that it is easier to rollover SUVs than non-SUV’s. From the case reviews of rollover crashes, it was seen that 61% of crashes occurred due to incapacitation, 26% due to driver inattention, and 9% due to failure to negotiate curve. It was seen that 79% of the drivers were ejected during the rollover. The main cause of rollover was tripping mostly on grass, which was seen in 80% of the cases.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    Rollover_Thesis_final.pdf 1.90 Mb 00:08:48 00:04:31 00:03:57 00:01:58 00:00:10