Scholarlink

Per country renewable water consumption is a widely analyzed indicator, reflecting sustainable water use at the community level. As a vital resource, water plays a central role in agriculture, nutrition, and public health dimensions collectively addressed by the One Water, One Health approach. Ensuring food safety and promoting health outcomes depend on the effective management of individual water consumption. To address the limitations of traditional single-model prediction methods, often marked by low precision and significant errors, we propose two advanced hybrid models: an ARIMA–LSTM (Long Short-Term Memory) model and a SARIMA–LSTM model adapted for Panel time series transformed into supervised Panel Times series. These models aim to enhance the accuracy of backward prediction of individual renewable water consumption, offering improved reliability for historical data estimation and decision-making processes.

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

Sustainable Development Backward Prediction Using a Hybrid Deep Learning Arima Model with an Auto-ARIMA Selection Criteria

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Belhassen Meftahi

摘要

Per country renewable water consumption is a widely analyzed indicator, reflecting sustainable water use at the community level. As a vital resource, water plays a central role in agriculture, nutrition, and public health dimensions collectively addressed by the One Water, One Health approach. Ensuring food safety and promoting health outcomes depend on the effective management of individual water consumption. To address the limitations of traditional single-model prediction methods, often marked by low precision and significant errors, we propose two advanced hybrid models: an ARIMA–LSTM (Long Short-Term Memory) model and a SARIMA–LSTM model adapted for Panel time series transformed into supervised Panel Times series. These models aim to enhance the accuracy of backward prediction of individual renewable water consumption, offering improved reliability for historical data estimation and decision-making processes.