The objective of this research is to develop a traffic accident forecasting model using traffic accident data in Taejeon metropolitan area from 1977 to 1994.
In this research, several forecasting models of a multiple regression model, a time-series A...
The objective of this research is to develop a traffic accident forecasting model using traffic accident data in Taejeon metropolitan area from 1977 to 1994.
In this research, several forecasting models of a multiple regression model, a time-series ARIMA model, and a Logistic curve model are developed. then Among them, the model which shows the most significance in forecasting accuracy is selected as the traffic accident forecasting model.
The results of this research are as followings.
1. The existing model such as Smeed's model which was developed for foreign countries shows some limitations in application to local unit such as Taejeon metropolitan area.
2. In multiple regression model, We developed two models by the types of traffic accident, and then established the model of more significance by regression diagnostic. As a result, the explanatory power of the forecasting models is 85% in the number of traffic accident forecasting model, 61% in the number of the death forecasting model and 93% in the number of the injury forecasting model respectively.
3. The traffic accidents in Taejeon metropolitan area is not explained by the time series ARIMA model well.
4. The limited number of each traffic accident types in the future is about 6,500 case in the number of traffic accidents, about 200 persons in number of the death, and about 6,800 persons in number of the injury according to trend of the number of traffic accident.