This thesis proposes a new spatial multivariate count model to jointly analyze the traffic crash-related counts of pedestrians and bicyclists by injury severity. The modeling framework is applied to predict injury counts at a Census tract level, based on crash data from Manhattan, New York. The results highlight the need to use a multivariate modeling system for the analysis of injury counts by road-user type and injury severity level, while also accommodating spatial dependence effects in injury counts. / text

Pedestrian injuries and fatalities due to motor vehicle crashes are a significant public health concern, and the urban campus of Georgia State University poses unique challenges to pedestrian safety issues. Previous studies of the built environment have link several features to increased pedestrian crash occurrences. Once identified, these features can be modified to create a healthier environment for pedestrians. This study examines the relationship between specific features of the built environment and pedestrian crash events. Environmental audits were conducted to collect information about the built environment around Georgia State campus, and pedestrian crash data was obtained from GDOT. Geographic Information Systems (GIS) was used to create a visual representation of this data in order to establish spatial relationships between the built environment and pedestrian crash events. Results show both positive and negative correlations between certain built environment features and pedestrian crashes. GIS was established as a useful tool for evaluating the spatial distribution and relationship between the built environment and pedestrian injury within a localized area, and provides a springboard for future research that seeks to study this association on a larger scale.

A study was done to explore the suitability of intersection and arterial collision prediction models based on traffic conflicts, generated using the Paramics microsimulation suite and the Surrogate Safety Assessment Model (SSAM). A linear regression model and a generalized linear model with a negative binomial error structure were explored to correlate conflicts to crash rates, as well as the conflict-based models suggested by SSAM. The model predictions were compared to volume-based predictions and historical data from Toronto, Ontario, Canada. The volume- based predictions were calculated using a negative binomial generalized linear model, fitted to the same arterial and intersection sets used to fit the conflict-based models. The results show the predictions generated by a conflict-based model were comparable for intersections, but poor for arterials.

A study was done to explore the suitability of intersection and arterial collision prediction models based on traffic conflicts, generated using the Paramics microsimulation suite and the Surrogate Safety Assessment Model (SSAM). A linear regression model and a generalized linear model with a negative binomial error structure were explored to correlate conflicts to crash rates, as well as the conflict-based models suggested by SSAM. The model predictions were compared to volume-based predictions and historical data from Toronto, Ontario, Canada. The volume- based predictions were calculated using a negative binomial generalized linear model, fitted to the same arterial and intersection sets used to fit the conflict-based models. The results show the predictions generated by a conflict-based model were comparable for intersections, but poor for arterials.

Software repositories such as source code repositories and bug repositories record information about the software development process. By analyzing the rich data available in software repositories, we can uncover interesting information. This information can be leveraged to guide software developers, or to automate software development activities. In this thesis we investigate two activities of the development process: selective code integration and grouping of field crash-reports, and use the information available in software repositories to improve each of the two activities. / Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2012-09-04 12:26:59.388

The objectives of this research were to explore ways to assess the safety performance of (predominantly two-lane) rural highways in New Zealand (NZ) and in particular identify driver/road/environmental factors affecting crashes on rural curves. Following a wide-ranging literature review, the Interactive Highway Safety Design Model (IHSDM) was identified as worthy of further investigation for adaptation to use in NZ. To help with this investigation, a comprehensive database was developed of road, traffic, crash and environmental data for all NZ State Highways, divided into variable-length road elements. A number of tasks were identified and undertaken to adapt IHSDM for general use here, including calibrating the Crash Prediction Module (CPM), developing a Design Policy file based on local agency standards, and developing an importing routine for NZ highway geometry and crash data. To assess the effectiveness of IHSDM for predicting the relative safety of rural road alignments, a series of tests were undertaken to confirm its appropriateness for use in NZ. These included “before and after” design consistency checks of a bridge replacement, a “before and after” crash comparison of a major highway realignment, and checks of actual versus predicted crash numbers along longer lengths of highway in varying terrain. These initial investigations have shown that IHSDM is a promising tool for safety and operational assessment of highway alignments (both existing and proposed) in NZ. Incorporating crash history data generally improves IHSDM’s accuracy in crash numbers, and appears to provide a better level of “local calibration” than by using sub-national (e.g. regional or terrain-specific) calibration parameters. Reported fatal/injury crash data generally provide more robust and precise measures than non-injury crashes. Correct specification of the extreme attributes of sub-standard elements (e.g. minimum radius, maximum roadside hazard) appears to be crucial to getting suitably accurate crash estimates on existing alignments. However, IHSDM’s current lack of consideration for bridges and inconsistent adjacent elements are notable omissions that limit the ability of the CPM to assess sub-standard existing routes with as much accuracy as well-designed newer alignments.

The overall aims of this PhD were to examine what the real effects of injury are on survivors of road crashes and to explore the methods used to assess these effects. Three studies were conducted for this thesis to explore the effects of injury using quality of life outcomes for survivors of road crashes. Study 1 was a prospective follow-up study of 70 road crash survivors with relatively 'minor' injuries. The aim of study 1 was to determine the effects of the crash or injury on quality of life for a 'minor' injury sample over a 1 year follow-up period. Study 2 was a prospective follow-up study of 50 'seriously' injured road crash survivors admitted to hospital as a result of their injuries. Study 3 incorporated the results of study 1 and study 2 to examine the societal burden of injury.

Intersection crashes accounted for 47 percent of the total number of crashes in the State of Georgia from 2000-2005, and as a location where crashes occur, represent the largest number of crash locations in the state. Federal legislation requires states to implement statewide safety plans to reduce fatalities, crashes, and improve safety. Intersections vary in different ways and there are individual factors that can cause an intersection to be safer or more dangerous than another. Acquiring better, uniform, and more updated information with regard to intersection crashes will enable transportation officials to prescribe policies for improving safety in an easier and more expedited manner. This thesis recommends a five-part program for intersection safety that will enable Georgia transportation officials to better analyze, identify, and implement countermeasures at intersections that are determined to be the most hazardous. The plan consists of: 1. Standardized Hazardous Intersection Identification Method 2. Statewide Public Involvement task force 3. Automated Police Crash Reporting Through Improved Technologies 4. Strategic Highway Safety Plan (SHSP) Intersection Safety Strategies (8 State Comparison) 5. Statewide Minimal Intersection Safety Equipment The five parts of this plan lead to a statewide standard method of analyzing intersections based on uniform collection methods and uniform equipment statewide.

The aim of this thesis was to examine the extent, and correlates, of self-regulation of driving behaviour among a sample of South Australian older drivers (aged 60 or more). The first of four studies was an analysis of official crash statistics in South Australia over a period of five years. The patterns of crash involvement for South Australian older drivers were found to resemble those reported in the literature for other jurisdictions. Lower levels of crash involvement for older drivers in difficult driving situations (peak hour, rain, darkness) were interpreted as indirect evidence for self-regulation of driving behaviour. The second study involved pilot testing a measure developed specifically for assessing the visual attention of older adults (the Computerised Visual Attention Test - CVAT). The CVAT assesses visual attention by measuring target detection and reaction time for central and peripheral stimuli, and in conditions requiring selective and divided attention. The third study involved assessing the test-retest reliability, construct validity and predictive validity of the CVAT. It was concluded that the CVAT is a reliable measure of abilities including, but not restricted to, attention, and that it is correlated with on-road driving ability. The fourth study involved an examination of the driving behaviour and attitudes of 104 drivers aged over 60, with avoidance of difficult driving situations providing an index of self-regulation. These drivers also completed a battery of tests measuring psychological factors, vision, physical functioning, various cognitive abilities, and attention (the CVAT). Ninety participants additionally completed an on-road assessment of driving ability. It was found that older drivers most often avoided reverse parallel parking and driving at night in the rain, while driving alone was avoided least often. Measures of visual attention, medication use and visual acuity were most predictive of levels of self-regulation, while poorer driving ability was only associated with avoidance of a small number of specific situations. Functional deficits related to poorer driving ability but not to self-regulation included poorer contrast sensitivity, speed of information processing and spatial memory. Such deficits could identify drivers who may need to restrict their driving more than they do. / Thesis (Ph.D.)--Department of Psychology, 2004.

Zugl.: München, Universiẗat der Bundeswehr, Diss., 2007.