Scholarlink

<p>This research explores how predictive analytics are used with the Facebook Prophet, Long Short-Term Memory (LSTM), and AutoRegressive Integrated Moving Average (ARIMA) models to enhance traffic management and accident prevention in smart cities. By leveraging traffic data collected from road sensors, the research aims to predict future traffic patterns, which improves road safety and traffic systems. The analysis looks at distinct traffic trends among diverse times of the day, weeks, and months, proving the effectiveness and efficiency of the models in accurately predicting traffic volume. The interpretations of the Prophet, ARIMA, and LSTM models achieved accuracy rates of 86.96%, 83.16%, and 86.64%, respectively. They are evaluated using other metrics such as mean relative absolute error (MAPE), root mean square error (RMSE), and mean absolute error (MAE). These results show that integrating predictive analytics into traffic management systems can transform them from reactive to proactive, which leads to optimized traffic flow, reduced congestion, and enhanced urban mobility. The study also addresses challenges related to data quality, real-time data processing, and the complexity of traffic patterns, which emphasizes the potential of AI-driven solutions in smart city initiatives and provides actionable insights for infrastructure planning and traffic management.</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

Integrating deep learning and statistical models for traffic prediction and accident prevention in smart cities

  • 期刊论文

  • 发表日期：

    2025年10月27日

查看全文
  • Yousef O. Sharrab,
  • Basma M. Irtahi,
  • Mohammad Ali H. Eljinini,
  • Izzat Alsmadi

摘要

This research explores how predictive analytics are used with the Facebook Prophet, Long Short-Term Memory (LSTM), and AutoRegressive Integrated Moving Average (ARIMA) models to enhance traffic management and accident prevention in smart cities. By leveraging traffic data collected from road sensors, the research aims to predict future traffic patterns, which improves road safety and traffic systems. The analysis looks at distinct traffic trends among diverse times of the day, weeks, and months, proving the effectiveness and efficiency of the models in accurately predicting traffic volume. The interpretations of the Prophet, ARIMA, and LSTM models achieved accuracy rates of 86.96%, 83.16%, and 86.64%, respectively. They are evaluated using other metrics such as mean relative absolute error (MAPE), root mean square error (RMSE), and mean absolute error (MAE). These results show that integrating predictive analytics into traffic management systems can transform them from reactive to proactive, which leads to optimized traffic flow, reduced congestion, and enhanced urban mobility. The study also addresses challenges related to data quality, real-time data processing, and the complexity of traffic patterns, which emphasizes the potential of AI-driven solutions in smart city initiatives and provides actionable insights for infrastructure planning and traffic management.