Development of Safety Performance Functions for SafetyAnalyst Applications in Florida

Abstract

During the time of my doctoral study at Florida International University, he helped me become a professional, responsible, and confident person. Dr. Gan's influence on my life has spanned many levels, at times as a professor, at other times, like a caring father.Dr. Gan taught me invaluable wisdom and problem-solving skills, both of which have been of great benefit to me during my studies, and will certainly continue to do so as I enter the engineering profession. He even took time out of his busy schedule to show me how to write professionally, from emails to technical papers. With Dr. Gan's help, I received many university, state, and international honors and awards. He has shown me that nothing is impossible if I simply believe and try my best. I am deeply honored to have been one of Dr. Gan's students. SafetyAnalyst implements the empirical Bayes (EB) method, which requires the use of Safety Performance Functions (SPFs). The system is equipped with a set of national default SPFs, and the software calibrates the default SPFs to represent the agency's safety performance. However, it is recommended that agencies generate agency-specific SPFs whenever possible. Many investigators support the view that the agency-specific SPFs represent the agency data better than the national default SPFs calibrated to agency data. Furthermore, it is believed that the crash trends in Florida are different from the states whose data were used to develop the national default SPFs.In this dissertation, Florida-specific SPFs were developed using the 2008 Roadway Characteristics Inventory (RCI) data and crash and traffic data from [2007][2008][2009][2010] for both total and fatal and injury (FI) crashes. The data were randomly divided into two sets, one for calibration (70% of the data) and another for validation (30% of the data). The negative binomial (NB) model was used to develop the Florida-specific SPFs vii for each of the subtypes of roadway segments, intersections and ramps, using the calibration data. Statistical goodness-of-fit tests were performed on the calibrated models, which were then validated using the validation data set. The results were compared in order to assess the transferability of the Florida-specific SPF models.The default SafetyAnalyst SPFs were calibrated to Florida data by adjusting the national default SPFs with local calibration factors.The performance of the Florida-specific SPFs and SafetyAnalyst default SPFs calibrated to Florida data were then compared using a number of methods, including visual plots and statistical goodness-of-fit tests. The plots of SPFs against the observed crash data were used to compare the prediction performance of the two models. Three goodness-of-fit tests,represented by the mean absolute deviance (MAD), the mean square prediction error (MSPE), and Freeman-Tukey R 2 (R 2 FT ), were also used for comparison in order to identify the better-fitting model. The results showed that Florida-specific SPFs yielded better prediction performance than the national default SPFs calibr...

Roadway Context Classification Approach for Developing Regional Safety Performance Functions for Florida Intersections

“…Both major AADT and minor AADT have a positive relationship with the total crash data, which means an increase in vehicles on the major or minor road is expected to increase the number of crashes. This positive relationship is consistent with the results of the SPF models developed in previous research that developed SPFs specific to Florida ( 4 ). The performance measure results for the rural 3-leg SPF developed in this previous research are mean squared prediction error (MSPE) = 187.14, mean squared error (MSE) = 192.4, and mean absolute deviance (MAD) = 8.21.…”

“…A previous study in Florida compared the results of locally calibrated SPFs with the default HSM SPFs. This study used four years of crash data (2007–2010) and developed SPFs for four types of intersections (rural and urban 3-leg and 4-leg) using a negative binomial regression model ( 4 ). Like the other studies, the results showed that the SPFs using Florida-specific calibration factors fit the data better than the models using HSM default values.…”

“…Intersections are common locations for crashes because of the presence of conflicting traffic movements. Traffic volume is a major factor in these crashes (higher volumes tend to result in more crashes), but recent studies have shown that the relationship between crash frequency and traffic volume is nonlinear (3)(4)(5)(6)(7)(8)(9). This is because several variables, such as roadway classification, area type, and geometric design affect the way drivers behave and the probability of crashes.…”

Prioritizing Highway Safety Manual’s crash prediction variables using boosted regression trees

“…Calibration factor is estimated as the ratio of the total number of observed crashes to the total number of predicted crashes calculated using the SPFs and CMFs provided in the HSM. The predictive models are most effective when calibrated to local conditions (Findley et al, 2012;Lu, 2013;Sun et al, 2006;Young and Park, 2013).…”

Improved Criteria for Estimating Calibration Factors for Highway Safety Manual (HSM) Applications

“…Calibration factor is estimated as the ratio of the total number of observed crashes to the total number of predicted crashes calculated using the SPFs and CMFs provided in the HSM. The predictive models are most effective when calibrated to local conditions (Findley et al, 2012;Lu, 2013;Sun et al, 2006;and Young and Park, 2013). …”