The present study intends to augment existing knowledge on pedestrian crossing behavior at non-intersection areas by analyzing crash data. The crash analysis was pivoted on the age and gender of an at-fault pedestrian who was involved in a crash when crossing the road at an undesignated midblock area. Pedestrian-vehicle crash information and narratives from 2016 to 2017 in Michigan State were used in the analysis. The frequent-based descriptive analysis, Chi-squared test, Ordered Probit Model, and Random Forest model were used to discern human, environmental and roadway factors associated with pedestrian crossing-related crashes at undesignated midblock areas. The results show that on average, a pedestrian was involved in a crash while crossing at an undesignated area about 130 feet from the nearest intersection. The most influential predictors of pedestrian fatalities while crossing at undesignated areas, as identified by the Random Forest model, were the lighting conditions (dark), pedestrian age, and traffic volume. Furthermore, the most important predictor of pedestrian fatalities were cases when a pedestrian was wearing dark clothing while crossing the road, as identified from crash narratives. The results of this study can be used by practitioners to identify locations other than intersections where there is a greater potential for pedestrian-vehicle conflicts. Also, the results can be used in efforts to educate the public on the importance of crossing at designated areas. (Author/publisher)
Abstract