Scholarlink

<p>This study examines the spatiotemporal trends of road traffic accidents leading to death and injury in the provinces of Mazandaran and Alborz over 41&#xa0;months. Space–time cube analysis within geographic information systems was applied, along with three forecasting models: Curve fit forecast (CFF), forest-based forecast, and exponential smoothing forecast. The results revealed that the CFF model provided the best performance, with RMSE values of 0.0030 for fatal accidents and 0.0003 for injury accidents. The forecasts indicated a 5.75% increase in hotspots for fatal accidents and a 34.3% rise in accident occurrences. Additionally, the forecasts showed a 4.82% increase in hotspots for injury-related accidents and a 30.93% expansion of high-risk areas. These findings are valuable for preventive planning, particularly in positioning emergency response stations closer to high-risk areas, thereby reducing response time and enhancing overall safety measures.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Forecasting spatiotemporal trends of road accidents injuries and fatalities using spatio-temporal cube models: a case study in Iran

  • 期刊论文

  • 发表日期：

    2025年6月25日

查看全文
  • Bahare Sadat Mousavi,
  • Meysam Argany,
  • Najme Neysani Samany,
  • Ahmad Soltani

摘要

This study examines the spatiotemporal trends of road traffic accidents leading to death and injury in the provinces of Mazandaran and Alborz over 41 months. Space–time cube analysis within geographic information systems was applied, along with three forecasting models: Curve fit forecast (CFF), forest-based forecast, and exponential smoothing forecast. The results revealed that the CFF model provided the best performance, with RMSE values of 0.0030 for fatal accidents and 0.0003 for injury accidents. The forecasts indicated a 5.75% increase in hotspots for fatal accidents and a 34.3% rise in accident occurrences. Additionally, the forecasts showed a 4.82% increase in hotspots for injury-related accidents and a 30.93% expansion of high-risk areas. These findings are valuable for preventive planning, particularly in positioning emergency response stations closer to high-risk areas, thereby reducing response time and enhancing overall safety measures.