고령 보행자 교통안전 개선 우선 지역 선정을 위한 방법론 개발

초록

This study aims to develop a systematic methodology for prioritizing regions that require elderly pedestrian traffic safety improvements at the city and county levels nationwide, in response to the increasing incidenceof elderly pedestrian accidents due to rapid population aging. Using traffic accident data involving elderly pedestrians and regional socio-economic indicators from 160 cities and counties across the nation, SafetyPerformance Functions (SPFs) were developed based on negative binomial and Conway–Maxwell–Poisson(Com-Poisson) models, with mixed-effects models applied to account for regional heterogeneity. Theanalysis identified elderly travel volume, residential area size, and annual average daily traffic as significant risk factors, while the number of elderly leisure and welfare facilities served as a protective factor. Modelfit and predictive performance were compared, and the Com-Poisson mixed-effects model demonstrated superior performance, leading to its selection as the final model. Subsequently, the Potential SafetyImprovement (PSI) index was calculated using the difference between model-predicted and observed accident counts, thereby identifying regions requiring urgent elderly pedestrian safety interventions. Theprioritized regions are expected to provide a valuable basis for the efficient allocation of limited resources and for the development of tailored safety measures that reflect local characteristics.

키워드