The study contributes to safety literature on transportation safety by employing copula based models for count frequency analysis at a macro-level. Most studies in the transportation safety area identify a single count variable (such as vehicular, pedestrian or bicycle crash counts) for a spatial unit and study the impact of exogenous variables. While the traditional count models perform adequately in the presence of a single count variable, it is necessary to modify these approaches to examine multiple dependent variables for each study unit. To that extent, the current research effort contributes to literature by developing two multivariate models based on copula methodology. First, a copula based bivariate negative binomial model for pedestrian and bicyclist crash frequency analysis is developed. Second, a multivariate negative binomial model for crashes involving non-motorized road users, passenger cars, vans, light trucks and heavy trucks is proposed. The proposed approaches also accommodate for potential heterogeneity (across zones) in the dependency structure. The formulated models are estimated using traffic crash count data at the Statewide Traffic Analysis Zone (STAZ) level for the state of Florida for the years 2010 through 2012. The STAZ level variables considered in our analysis include exposure measures, socio-economic characteristics, road network characteristics and land use attributes. A policy analysis is also conducted along with a representation of hotspot identification to illustrate the applicability of the proposed model for planning purposes. The development of such spatial profiles will allow planners to identify high risk zones for screening and treatment purposes.

Fog along roadways is a dangerous hazard that leads to crashes resulting from limited visibility. Low visibility gives drivers less time to react to potential obstacles that can suddenly appear and require immediate action. To solve this issue, early warning systems involving Dynamic Message Signs or other types of devices are used to alert drivers of the impending visibility condition so that they are prepared. This research focuses on testing the effectiveness of one form of warning systems to investigate how it impacts driver behavior in foggy conditions. To accomplish this objective, a simulation study is developed to test variables of interest including: Roadway Type, Fog Level, DMS Presence, Beacon Presence, Traffic Volume, and DMS Message Provided. Using a factorial design, 24 scenarios are created by randomizing the variables listed using statistical software to be tested on 72 volunteer participants. Using a NADS MiniSim Driving Simulator, the participants driving behavior is recorded including speed and breaking behavior under an initial clear condition followed by a reduced visibility fog condition. From demographics, drivers age 35 and over consistently showed a higher likelihood of speed reduction between clear and fog conditions with overall reduction increasing with age. This is seen when looking at the mean change in speed based on driver age where young drivers (18-25 yrs) reduced speeds by 7MPH, older drivers (35-45 yrs) reduced by 12MPH, and elder drivers (65+ yrs) reduced by 17MPH. The more often a person drove and those that were educated at a graduate level also showed a higher chance of speed reductions. This demonstrates the impact of experience and exposure to driving performance under reduced visibility conditions. Those who recently drove under fog conditions or learned to drive in Florida were found to be less likely to reduce their speeds when entering the fog. This is attributed to these drivers being confident or familiar with the environment resulting in risky driving behavior. For the scenario variables, it is determined that the type of roadway a driver travels plays a major role in how much speed reduction occurs and thus how much a driver decelerates when entering a low visibility environment. On average, drivers traversed the fog zone at 50MPH with the lowest travel speed being 30MPH. Since the speed limit on the freeway is 5MPH higher than the arterial, drivers' traveling along this road are noted to decelerate at higher rates to achieve this target speed. Additionally, DMS presence and message also provided an impact on the drivers' choice to decelerate and reduce travel speed within the fog condition. Under the most severe conditions, the probability of a driver reducing speed increases as the number of DMS present increases. Additionally, when a DMS presents a warning and specifies the action that a driver should take, in this case 'reduce speed,' greater speed reductions and decelerations are observed and are more likely to occur. Interestingly the number of DMS did not have a significant impact on driver behavior under every fog condition like the message presented did except in the most severe fog condition. Taking into account that 33% of drivers did not accurately remember the number of DMS encountered it can be concluded that the warning message itself is the most important aspect of the early warning system. This indicates that drivers accurately remember being directed to reduce speed whether they are given the advisement once or multiple times based on the number of DMS present. Further research into how the warning message is presented or worded could provide additional insight into the impact it can have on driver behavior. Since it is observed that drivers acknowledge the 'reduce speed' advisement, it is likely that specifying a specific speed limit could also warrant driver obedience. Additional testing and observation of driver reaction to larger traffic volumes and situations within the fog would also allow for further analysis of driver behavior under reduced visibility and the impact the early warning system has on their behavior.

Toll plazas are becoming an essential part of the highway system, especially within the state of Florida. Many crashes reported on highways occur at toll plazas. A primary reason for many vehicle collisions happening at these facilities is the fact that each toll plaza agency has different design, signage and marking criteria. This, in turn, causes driver confusion and possible last minute weaving maneuvers. Even though the varying design of toll plazas is a clear highway safety factor, research in the field is very limited but expanding. This study focuses on one toll plaza, in particular the Dean Mainline Toll Plaza, located in Orlando, Florida. The toll plaza is located directly between two roads that are in close proximity of each other. Because of this, the toll plaza is very close to the on- and off- ramps, which can be even more confusing and stressful for a driver entering or leaving the highway. The purpose of this study is to evaluate the safety and efficiency of the Dean Mainline Toll Plaza in order to make recommendations to improve or maintain the current toll plaza design, as well as potentially contribute to a nationally set design standard for toll plazas. Using the NADS miniSimTM Simulator, 72 subjects were recruited, and each subject was asked to drive 3 scenarios that were randomly selected from a pool of 24 scenarios. The following factors were changed in order to study the driver's behavior: signage and their location, pavement markings, distances between the toll plaza and ramps, and traffic conditions. All of these factors were altered and observed on five of the eight possible routes than can be taken through the toll plaza. The subjects were asked to complete questionnaires before and after all of the scenarios, as well as in between each driving scenario. These questionnaires included demographic characteristics, such as age, education, income, E-PASS ownership, etc. The data that were collected by the driving simulator and questionnaires were analyzed by ANOVA and multinomial logistic regression models. A positive relationship was found between non-urgent lane changing and the current real-world sign conditions prior to the toll plaza. Relationships were also found between the subjects' speed in various locations and signage before the toll plaza and segment length after the toll plaza. Along with specified recommendations for future research in toll plaza safety, recommendations for the Dean Mainline Toll Plaza include maintaining the current signs and pavement markings, as they were found to be beneficial in drivers performing safe lane changing maneuvers.

For the past several decades, pedestrian safety has been an oncoming issue that has thrown the area of transportation engineering into a frenzy. Pedestrian safety has become predominantly one of the leading causes of fatalities in traffic accidents. Florida has been reported as one of the leading states in pedestrian fatalities with 2.56 fatality rate per 100,000 population and about 20 percent of all traffic fatalities in the state of Florida. Nonetheless, as research is being done and hypotheses are being calibrated and produced, there has to be a way of measuring and determining the number of pedestrian-to-vehicle conflicts without having to yet apply the system on the field without further validation. Moreover, pedestrian-to-vehicle conflicts have been a rising issue in correlation to the pedestrian fatalities. The fact that the highway safety manual has limited information about crash modification functions for pedestrian and that pedestrian fatality is a rare event, it is worthwhile identifying and adopting surrogate safety measures for pedestrian. Thus, having the capability to analyze various surrogate safety measures within the confines of micro simulation would be a great contribution to real-world application. As a result, the purpose of this thesis is to determine the feasibility of using micro simulation to assess safety of pedestrian crossings using specifically VISSIM and SSAM. During this study, a great deal of data extraction was taken from videotapes collected at nine various intersections, each with its own environmental and geometrical factors. Various parameters were taken from the different sites in order to calibrate and validate VISSIM and SSAM. The parameters included traffic and pedestrian volumes, walking speeds, crossing times, signal timings, and pedestrian-to-vehicle conflicts. During this study, an extensive amount of analysis testing was done in order to obtain the optimum threshold within various combinations of thresholds that would define the pedestrian-to-vehicle conflicts. The analysis was initiated for the time to collision (TTC) and post encroachment time (P.E.T) thresholds. This is done so that the typical scenario of an intersection can be analyzed and comparisons can be made efficiently between observed and simulated conflicts. There were 55 combinations of TTC and PET thresholds produced that were also statistically calculated using the mean absolute percent error (MAPE) in order to determine the most efficient threshold for all 9 intersections. Calibration also was done for parameters in VISSIM that included the safety distance factor (SDF) and the Add-stop distance to assess the sensitivity of these parameters in computing the number of pedestrian-to-vehicle conflicts. These thresholds and factors were used for further validation and assessment of the feasibility of the SSAM and VISSIM model. Data results displayed that the simulated conflicts and the observed conflicts illustrated reasonable correlation. However, even with the feasibility of VISSIM and SSAM being validated, there still are questions that arise pertaining to whether VISSIM and other micro simulation can assess real-world driver behavior and the unpredictability of driver maneuvering. More research with more intersections are recommended to be done.

Thesis (PhD)--University of South Australia, 1996

Thesis (PhD)--University of South Australia, 1996

Variable speed limit (VSL) systems are important active traffic management tools that are being deployed across the U.S. and indeed around the world for relieving congestion and improving safety. Oregon’s first variable advisory speed limit signs were activated along Oregon Highway 217 in the summer of 2014. The variable advisory speed system is responsive to both congestion and weather conditions. This seven-mile corridor stretches around Western Portland and has suffered from high crash rates and peak period congestion in the past. VSL systems are often deployed to address safety, mobility and sustainability related performance. This research seeks to determine whether the newly implemented variable advisory speed limit system has had measurable impacts on traffic safety and what the scale of the impact has been. The research utilizes a before-after crash analysis with three years of data prior to implementation and around 16 months after. Statistical analysis using an Empirical Bayes (EB) approach will aim to separate the direct impacts of the variable advisory speed limit signs from the long term trends on the highway. In addition, the analysis corrects for the changes in traffic volumes over the study period. Three data sources will be utilized including Washington County 911 call data, Oregon incident reports, and official Oregon Department of Transportation crash data reports. The analysis results are compared between data sources to determine the reliability of 911 call data as a proxy for crash statistics. The conclusions should be able to provide an indication of whether variable advisory speed limits can provide increased safety along high crash corridors.

The goal of this study is to investigate the predictive ability of less data intensive but widely accepted methods to estimate mobility and reliability measures. Mobility is a relatively mature concept in the traffic engineering field. Therefore, many mobility measure estimation methods are already available and widely accepted among practitioners and researchers. However, each method has their inherent weakness, particularly when they are applied and compared with real-world data. For instances, Bureau of Public Roads (BPR) Curves are very popular in static route choice assignment, as part of demand forecasting models, but it is often criticized for underperforming in congested traffic conditions where demand exceeds capacity. This study applied five mobility estimation methods (BPR Curve, Akcelic Function, Florida State University (FSU) Regression Model, Queuing Theory, and Highway Capacity Manual (HCM) Facility Procedures) for different facility types (i.e. Freeway and Arterial) and time periods (AM Peak, Mid-Day, PM Peak). The study findings indicate that the methods were able to accurately predict mobility measures (e.g. speed and travel time) on freeways, particularly when there was no congestion and the volume was less than the capacity. In the presence of congestion, none of the mobility estimation methods predicted mobility measures closer to the real-world measure. However, compared with the other prediction models, the HCM procedure method was able to predict mobility measures better. On arterials, the mobility measure predictions were not close to the real-world measurements, not even in the uncongested periods (i.e. AM Peak and Mid-Day). However, the predictions are relatively better in the AM and Mid-Day periods that have lower volume/capacity ration compared to the PM Peak period. To estimate reliability measures, the study applied three products from the Second Strategic Highway Research Program (SHRP2) projects (Project Number L03, L07, and C11) to estimate three reliability measures; the 80th percentile travel time index, 90th percentile travel time index, and 95th percentile travel time index. A major distinction between mobility estimation process and reliability estimation process lies in the fact that mobility can be estimated for any particular day, but reliability estimation requires a full year of data. Inclusion of incident days and weather condition are another important consideration for reliability measurements. The study found that SHRP2 products predicted reliability measures reasonably well for freeways for all time periods (except C11 in the PM Peak). On arterials, the reliability predictions were not close to the real-world measure, although the differences were not as drastic as seen in the case of arterial mobility measures.

This study develops, analyzes, and a­­­­­pplies transit-system-specific regression tree models that identify and prioritize transit system improvements through analysis and application of ridership, Census, routing and scheduling, and transit stop characteristic data. Regression trees identify and rank independent variables that split dependent variable datasets into meaningful subsets according to significant relationships with independent variable datasets, and regression tree models can be used to identify and prioritize transit system improvements. In this study, ridership datatypes are the dependent variables (i.e., boardings and alightings) and Census, routing and scheduling, and transit stop characteristic datatypes are the independent variables. Data associated with the San Luis Obispo Regional Transit Authority (RTA) is the basis of this study. The literature review for this study identified no other studies that use regression trees to identify and/or prioritize transit system improvements. The analysis method herein can help identify and prioritize improvements to any transit system. The findings of this study may be applicable to other transit systems if assumptions can be made about the similarity of other systems to the San Luis Obispo Regional Transit Authority system. Relationships between transit ridership and independent variables that may be effective predictors of transit ridership are evaluated in this study. Traditional independent variables used to forecast transit ridership include population and employment densities, land use types, income distributions, service frequencies, and transit stop accessibility; other independent variables that may be significant predictors of transit ridership include transit stop amenities, characteristics, and connecting and nearby infrastructure. Ridership data needed for the analysis presented in this study can be obtained from transit agencies. Census data needed for the analysis presented in this study is available through the United States Census Bureau. Routing and scheduling data needed for the analysis presented in this study can be extracted from local transit system schedules. Transit stop characteristic data needed for the analysis presented in this study can be gathered by using a survey instrument during field-visits. The regression tree models developed in this study show a positive relationship in the RTA system between transit ridership and population density (specifically Asian and twenty to twenty-four years old residential population densities), the number of trips serving transit stops, and transit stop characteristics (specifically the presence of a trash can). According to these findings, this study offers recommendations for improvements to RTA’s transit system and marketing and planning strategies. More general conclusions that could be applicable to more transit systems could be drawn if the analysis method used in this study were performed with more and/or larger datasets (e.g., other transit agency, regional, statewide, national, and/or global datasets) comprised of more robust, accurate, and precise datatypes, and this concept is the basis for the future work recommended by this study.

Many real-world problems could be formulated in a way to fit the necessary form for discrete optimization. Discrete optimization problems can be solved by numerous different techniques that have developed over time. Some of the techniques provide optimal solution(s) to the problem and some of them give "good enough" solution(s). The fundamental reason for developing techniques capable of producing solutions that are not necessarily optimal is the fact that many discrete optimization problems are NP-complete. Metaheuristic algorithms are a common name for a set of general-purpose techniques developed to provide solution(s) to the problems associated with discrete optimization. Mostly the techniques are based on natural metaphors. Discrete optimization could be applied to countless problems in transportation engineering. Recently, researchers started studying the behavior of social insects (ants) in an attempt to use the swarm intelligence concept to develop artificial systems with the ability to search a problem's solution space in a way that is similar to the foraging search by a colony of social insects. The development of artificial systems does not entail the complete imitation of natural systems, but explores them in search of ideas for modeling. This research is partially devoted to the development of a new system based on the foraging behavior of bee colonies — Bee System. The Bee System was tested through many instances of the Traveling Salesman Problem. Many transportation-engineering problems, besides being of combinatorial nature, are characterized by uncertainty. In order to address these problems, the second part of the research is devoted to development of the algorithms that combine the existing results in the area of swarm intelligence (The Ant System) and approximate reasoning. The proposed approach — Fuzzy Ant System is tested on the following two examples: Stochastic Vehicle Routing Problem and Schedule Synchronization in Public Transit. / Ph. D.