Identification of contributing factors and accuracy of fault prediction using various sources of fatal pedestrian crash data in Florida

The principal focus of this research was to develop improved methods for analyzing pedestrian crash records using data from various sources. A detailed case study review of 353 fatal pedestrian crashes that occurred in Florida, primarily in the year 2000, was conducted. Data sources reviewed and used to construct a crash database for analysis included state transportation records, traffic crash reports, traffic homicide investigative report narratives, diagrams, photographs, select accident reconstructions and site visit notes. Custom pedestrian crash types for Florida conditions were developed. Salient factors contributing to fatal pedestrian crashes in Florida and trends among predominant pedestrian crash types were identified. New algorithms to predict fault in pedestrian crashes based on a combination of additional data and more accurate data were developed. An alternative to the current fault prediction algorithm used by the State of Florida Safety Office that uses the same source data was also developed and evaluated.; Analysis of the data indicated that the most significant causes of pedestrian crashes were pedestrian behavior, alcohol use by pedestrians and drivers, poor pedestrian visibility at night coupled with violation of driver expectation, and lack of compliance with state laws. Some form of pedestrian behavior was the primary contributing factor in over three-fourths of the pedestrian crashes reviewed. Alcohol use, by either the pedestrian or driver, was determined as the primary factor in 45% of cases. Where alcohol use was determinable, 69% of pedestrians crossing not in crosswalks were under the influence. Dark conditions or insufficient lighting was a contributing factor in 60% of cases. In nearly half of the non-crosswalk crossing cases, pedestrians were attempting to cross the road within 600 feet of a traffic signal, many times violating driver expectation and right-of-way. Other factors were detected within a specific type of pedestrian crash; for instance, in half of the 15% of cases that occurred on limited access facilities, a significant contributing factor was the former occupant having exited a disabled vehicle. In 57% of the cases where a pedestrian was walking along the roadway, there was not a sidewalk for the pedestrian to use. A suggested list of countermeasures is provided.; 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.