Development and Application of Accident Prediction Model for Railroad At-Grade Crossings

철도건널목의 사고예측모형 개발에 관한 연구

Rail crossings pose special safety concerns for modern railroad operation with faster trains. More than ninety percent of train operation-related accidents occurs on at-grade crossings. Surest countermeasure for this safety hazard is to eliminate at-grade crossings by constructing over/under pass or by closing them. These eliminations usually require substantial amount of investment and/or heavy public protest from those affected by them. Thorough and objective analysis are usually required, and valid accident prediction models are essential to the process. This paper developed an accident prediction model for Korean at-grade crossings. The model utilized many important factors such as guide personnel, highway traffic, train frequency, train sight distance, and number of tracks. Developed model was validated with actual accident data.