Impact of COVID-19 on Household Waste Generation: An Empirical Study Using Machine Learning-based Neural Network

Sung Won Choi, Jai Young Lee

Research output: Contribution to journalArticlepeer-review

Abstract

An accurate forecasting of waste generation is important to establish an efficient waste management. Although the patterns of waste generation are changing due to the COVID-19 pandemic, which started in 2020, there has been no accurate analysis and evaluation. Therefore, this study aims to assess the effects of COVID-19 on household waste generation using machine learning-based neural networks (ANN, LSTM, RNN, and GRU). This study partitioned data into training data (January 2017 – June 2021) and verification data (July 2021 – May 2022) and predicted future waste generation for June 2022 – December 2028. The results showed excellent prediction performance except for high-volatility recyclable waste. Forecasting based on household waste not by type can guarantee predictive performance. In addition, the impact of COVID-19 on household waste generation was a 4.65% increase in garbage waste, a 9.4% increase in food waste, and a 15.53% decrease in recyclable waste.

Original languageEnglish
Pages (from-to)12-20
Number of pages9
JournalJournal of Korea Society of Waste Management
Volume40
Issue number1
DOIs
StatePublished - 2023

Keywords

  • Forecasting model
  • Household waste generation
  • Machine learning
  • Neural Network

Fingerprint

Dive into the research topics of 'Impact of COVID-19 on Household Waste Generation: An Empirical Study Using Machine Learning-based Neural Network'. Together they form a unique fingerprint.

Cite this