Analysis Of Characteristic Factors Contributing To Hit-And-Run

Hit-and-run is a special form of accidents, which compared to other types of accidents, the investigation and handling of such accidents could be more difficult, and the emergency agency also could not work properly due to fail to realizing the accident in time. Collecting hit-and-run accident data is also very hard:the percentage of missing data is higher than that of the other types of accident’s data, which also caused some difficulties to the research and the prevention. The studying of this type of accident’s characteristic factors would be helpful to reduce the number of hit-and-run accidents, and also has important theoretical significance and plays a practical guidance role. Currently, the researching of hit-and-run accidents have certain progresses and have achieved many results from various perspectives.In this thesis, we developed an accident characteristic factor identification from analyzing accident data. At the beginning part of this thesis, we introduced the situation of researching in the field of hit-and-run, and proved the necessity of researching on this kind of accident. Based on Michigan’s accident data in2009, we found out the error data’s existing form in hit-and-run database, the error types of accident data were also classified. By using data processing techniques, mainly the indicator method, we reduced the impact of error data to a lower level. Then we found out representative factors, and introduced into a logistic regression model. Subsequently, using the SPSS statistical analysis software estimated the regression models’parameters with the various types of factors (environment, driver and vehicle) and quantified the results of the various factors for the hit-and-run probability. Finally, by model testing provided the goodness-of fit on different model for the hit-and-run accident data. That this paper will provide a certain practical significance for the reducing the probability of hit-and-run occurrences, and plays a technical reference role to the traffic management decision-makers.