Binary logistic regression was shown to be an effective technique for modeling fault in pedestrian crashes. Results were used to classify fault, and identify additional factors that influence fault. Logistic models correctly classified fault in anywhere from 84% to 97% of cases as compared to the existing Florida Department of Transportation (FDOT) algorithm, which only classified fault correctly in 56% to 58% of the same cases. Improvements in classification accuracy were shown to stem from two sources. The first was the use of abundant data, which took advantage of data augmentation; a process by which additional fields of data were made available for investigation. The second source of improvements was from the use of more accurate data; data which had undergone quality enhancements to correct errors, and to fill in incomplete or missing information. Both the improvements in quality and quantity of data came from using additional data sources and manual case reviews.

Driver workload increases on horizontal curves due to more complicated navigation compared to navigation on straight roadway sections. Although only a small portion of roadways are horizontal curve sections, approximately 25% of all fatal highway crashes occur at horizontal curve sections. According to the Fatality Analysis Reporting System (FARS) database, fatalities associated with horizontal curves were more than 25% during last years from 2008 to 2014, reinforcing that investigation of horizontal curve crashes and corresponding safety improvements are crucial study topics within the field of transportation safety. Improved safety of horizontal curve sections of rural transportation networks can contribute to reduced crash severities and frequencies. Statistical methods can be utilized to develop crash prediction models in order to estimate crashes at horizontal curves and identify contributing factors to crash occurrences, thereby correlating to the primary objectives of this research project. Primary data analysis for 221 randomly selected horizontal curves on undivided two-lane two-way highways with Poisson regression method revealed that annual average daily traffic (AADT), heavy vehicle percentage, degree of curvature, and difference between posted and advisory speeds affect crash occurrence at horizontal curves. The data, however, were relatively overdispersed, so the negative binomial (NB) regression method was utilized. Results indicated that AADT, heavy vehicle percentage, degree of curvature, and long tangent length significantly affect crash occurrence at horizontal curve sections. A new dataset consisted of geometric and traffic data of 5,334 horizontal curves on the entire state transportation network including undivided and divided highways provided by Kansas Department of Transportation (KDOT) Traffic Safety Section as well as crash data from the Kansas Crash and Analysis Reporting System (KCARS) database were used to analyze the single vehicle (SV) crashes. An R software package was used to write a code and combine required information from aforementioned databases and create the dataset for 5,334 horizontal curves on the entire state transportation network. Eighty percent of crashes including 4,267 horizontal curves were randomly selected for data analysis and remaining 20% horizontal curves (1,067 curves) were used for data validation. Since the results of the Poisson regression model showed overdispersion of crash data and many horizontal curves had zero crashes during the study period from 2010 to 2014, NB, zero-inflated Poisson (ZIP), and zero-inflated negative binomial (ZINB) methods were used for data analysis. Total number of crashes and severe crashes were analyzed with the selected methods. Results of data analysis revealed that AADT, heavy vehicle percentage, curve length, degree of curvature, posted speed, difference between posted and advisory speed, and international roughness index influenced single vehicle crashes at 4,267 randomly selected horizontal curves for data analysis. Also, AADT, degree of curvature, heavy vehicle percentage, posted speed, being a divided roadway, difference between posted and advisory speeds, and shoulder width significantly influenced severe crash occurrence at selected horizontal curves. The goodness-of-fit criteria showed that the ZINB model more accurately predicted crash numbers for all crash groups at the selected horizontal curve sections. A total of 1,067 horizontal curves were used for data validation, and the observed and predicted crashes were compared for all crash groups and data analysis methods. Results of data validation showed that ZINB models for total crashes and severe crashes more accurately predicted crashes at horizontal curves. This study also investigated the effect of speed limit change on horizontal curve crashes on K-5 highway in Leavenworth County, Kansas. A statistical t-test proved that crash data from years 2006 to 2012 showed only significant reduction in equivalent property damage only (EPDO) crash rate for adverse weather condition at 5% significance level due to speed limit reduction in June 2009. However, the changes in vehicles speeds after speed limit change and other information such as changes in surface pavement condition were not available. According to the results of data analysis for 221 selected horizontal curves on undivided two-lane highways, tangent section length significantly influenced total number of crashes. Therefore, providing more information about upcoming changes in horizontal alignment of the roadway via doubling up warning sings, using bigger sings, using materials with higher retroreflectivity, or flashing beacons were recommended for horizontal curves with long tangent section lengths and high number of crashes. Also, presence of rumble strips and wider shoulders significantly and negatively influenced severe SV crashes at horizontal curve sections; therefore, implementing rumble strips and widening shoulders for horizontal curves with high number of severe SV crashes were recommended.

This study applied tree-based machine learning methods to investigate the contributing factors to both crash frequency and injury severity in vehicle-pedestrian crash events. Vehicle and roadway characteristics, driver and pedestrian attributes, traffic controls and land use conditions, transit provision and weather conditions are used as covariates to predict pedestrian crash frequencies (by roadway segment) and injury severity levels (for pedestrians struck by vehicles on public roadways). In both cases, tree-based models offered significantly more prediction accuracy than traditional statistical models (using negative binomial and ordered probit specifications, with the same covariates). The tree-based models also offer valuable interpretability through the regression tree graph itself (with clear branching based on variable cut-points), variable importance plots (for each covariate), and partial dependence plots to help analysts understand the relationship between contributing factors and the target variable (count or severity). Average daily vehicle-miles travelled (DVMT) on each road segment, population density, segment length, census tract-level job density, distance from nearest K-12 school, transit stop density, and segment speed limits were estimated to be the top contributing factors for increasing pedestrian crash counts. DVMT has been found as the single most responsible factor for vehicle-pedestrian crashes and in a way representing pedestrian exposure to such situations. In terms of predicting injury outcomes, intoxication of the pedestrian and/or driver, higher speed limits at the site, crash location not being in the traffic way, older pedestrian, interstate highway locations, and dark and unlit conditions were predictors for more severe outcomes. Importantly, if the surrounding urban area’s population is reasonably high (over 25,000 persons), the probability of the pedestrian dying falls significantly, which supports the ‘safety in numbers’ idea, for more people available to help save the crash victims, or drivers going more slowly due to crowded conditions, closer hospitals, and so on. While few crash studies have included land use variables and local demographics, including proximity to schools, hospitals, and transit stops, population and jobs density variables appeared to add to crash counts and severity for pedestrians, though this is moderated by the 25,000-population threshold and distance variables

Reduction of crash occurrence on the various roadway locations (mid-block segments; signalized intersections; un-signalized intersections) and the mitigation of injury severity in the event of a crash are the major concerns of transportation safety engineers. Multi lane arterial roadways (excluding freeways and expressways) account for forty-three percent of fatal crashes in the state of Florida. Significant contributing causes fall under the broad categories of aggressive driver behavior; adverse weather and environmental conditions; and roadway geometric and traffic factors. The objective of this research was the implementation of innovative, state-of-the-art analytical methods to identify the contributing factors for crashes and injury severity. Advances in computational methods render the use of modern statistical and machine learning algorithms. Even though most of the contributing factors are known a-priori, advanced methods unearth changing trends. Heuristic evolutionary processes such as genetic programming; sophisticated data mining methods like conditional inference tree; and mathematical treatments in the form of sensitivity analyses outline the major contributions in this research. Application of traditional statistical methods like simultaneous ordered probit models, identification and resolution of crash data problems are also key aspects of this study. In order to eliminate the use of unrealistic uniform intersection influence radius of 250 ft, heuristic rules were developed for assigning crashes to roadway segments, signalized intersection and access points using parameters, such as 'site location', 'traffic control' and node information. Use of Conditional Inference Forest instead of Classification and Regression Tree to identify variables of significance for injury severity analysis removed the bias towards the selection of continuous variable or variables with large number of categories. For the injury severity analysis of crashes on highways, the corridors were clustered into four optimum groups. The optimum number of clusters was found using Partitioning around Medoids algorithm. Concepts of evolutionary biology like crossover and mutation were implemented to develop models for classification and regression analyses based on the highest hit rate and minimum error rate, respectively. Low crossover rate and higher mutation reduces the chances of genetic drift and brings in novelty to the model development process. Annual daily traffic; friction coefficient of pavements; on-street parking; curbed medians; surface and shoulder widths; alcohol / drug usage are some of the significant factors that played a role in both crash occurrence and injury severities. Relative sensitivity analyses were used to identify the effect of continuous variables on the variation of crash counts. This study improved the understanding of the significant factors that could play an important role in designing better safety countermeasures on multi lane highways, and hence enhance their safety by reducing the frequency of crashes and severity of injuries. Educating young people about the abuses of alcohol and drugs specifically at high schools and colleges could potentially lead to lower driver aggression. Removal of on-street parking from high speed arterials unilaterally could result in likely drop in the number of crashes. Widening of shoulders could give greater maneuvering space for the drivers. Improving pavement conditions for better friction coefficient will lead to improved crash recovery. Addition of lanes to alleviate problems arising out of increased ADT and restriction of trucks to the slower right lanes on the highways would not only reduce the crash occurrences but also resulted in lower injury severity levels.

In 2004, a total of 95,020 vehicle crashes occurred on highways under the jurisdiction of the Virginia Department of Transportation (VDOT). Of these, 39,847 crashes occurred on primary highways, and 345 of these were fatal crashes. VDOT's traffic engineers continue to place increasing emphasis on identifying causal factors for crashes to enhance the selection of appropriate and effective countermeasures. The purpose of this study was to identify causal factors and appropriate countermeasures for crashes occurring at high-risk locations on multilane primary highways from 2001 through 2006. These high-risk locations were identified by Fontaine and Reed (2006) in a VDOT safety corridor study. A total of 365 sites, 1 to 2 mi in length, were used in the study. The statewide sites were located on rural and urban highways with divided, undivided, and traversable medians, with about 40 sites per VDOT district. Crash data were extracted from police crash reports, and geometric data were collected through site visits. Operational data were collected using VDOT's resources. The analysis involved more than 34,000 crashes and was conducted using fault tree analysis and generalized linear modeling. The fault tree analysis was used to determine the critical fault path based on the probability of an event occurring. Individual fault trees were constructed for each collision type and for each highway classification. The generalized linear models were developed for different highway classifications: urban divided, urban undivided, urban traversable (central lanes that can be used for left turns in both directions), and rural divided highways. Models were developed for rear-end crashes and total crashes, and separate models were developed for injury crashes, property damage only (PDO) crashes, and injury + PDO crashes. Appropriate potential countermeasures were then identified based on the significant causal factors identified in the models. The results indicated that rear-end crashes were the predominant type of crash, representing 56% of all PDO crashes on urban divided highways and 37% of all PDO crashes on rural divided highways. Implementing the recommended countermeasures for total, rear-end, and angle crashes for different assumed levels of rehabilitation is expected to result in a crash reduction of up to about 40% depending on the site and level of rehabilitation undertaken. A benefit/cost analysis showed that the benefit/cost ratios were higher than 1 for all levels of countermeasure implementation.