Scholarlink

Establishing a reliable model for predicting carbon emissions in construction is advantageous for achieving high-quality development in this sector and promoting sustainable societal progress. This study adopts two representative time series analysis methods, namely autoregressive integrated moving average (ARIMA) model and grey model (GM), to construct a combined model for predicting carbon emissions. The combined model not only captures the trends and seasonal variations in time series data, but also handles nonlinear features in data. Using the historical data from Jiangsu Province’s construction industry from 2006 to 2022, we apply the combined model to forecast carbon emissions. Experimental results demonstrate that the combined approach significantly improves prediction accuracy compared to using either method individually. This study offers new technical support for monitoring and forecasting carbon emissions in construction and provides a scientific basis for formulating emission reduction policies, thereby propelling the construction industry towards more sustainable development.

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

Building Carbon Emission Prediction Model Based on Combination of Grey Forecasting and ARIMA

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shuai Huang,
  • Caiqian Wang,
  • Qian Hu,
  • Qing Luo

摘要

Establishing a reliable model for predicting carbon emissions in construction is advantageous for achieving high-quality development in this sector and promoting sustainable societal progress. This study adopts two representative time series analysis methods, namely autoregressive integrated moving average (ARIMA) model and grey model (GM), to construct a combined model for predicting carbon emissions. The combined model not only captures the trends and seasonal variations in time series data, but also handles nonlinear features in data. Using the historical data from Jiangsu Province’s construction industry from 2006 to 2022, we apply the combined model to forecast carbon emissions. Experimental results demonstrate that the combined approach significantly improves prediction accuracy compared to using either method individually. This study offers new technical support for monitoring and forecasting carbon emissions in construction and provides a scientific basis for formulating emission reduction policies, thereby propelling the construction industry towards more sustainable development.