Predicting municipal solid waste generation using a multi-city machine learning model

Predicting municipal solid waste generation using a multi-city machine learning model

None. Credit: Wenjing Lu, Weizhong Huo, Huwanbieke Gulina, Chao Pan.

The ever-increasing production of solid waste has been threatening the natural environment and human safety for several years. With increasing urbanization around the world, municipal solid waste (MSW) has increased dramatically. Integrated MSW management is an efficient method, but accurate prediction of MSW generation is a complex problem. Some traditional prediction models (multivariate linear regression model, time series analysis model, etc.) are successful using simple methods, but they usually select a basic mathematical model in advance, which limits the ability to truly reflect the characteristics of MSW.

High-accuracy machine prediction models, which can obtain complex new data and exploit it in depth, are increasingly being used to create short-, medium-, and long-term predictions for MSW generation. Among them, algorithms such as artificial neural network (ANN), support vector machine (SVM), and gradient boosting regression tree (GBRT) were used to predict the generation of MSW. However, the lack of a high-precision model based on large-scale data collection and a wide range of influencing variables limits the broad applicability of the model.

To meet the needs of the large-scale comprehensive treatment and realize the short-term MSW generation prediction, Professor Weijing Lu of Tinghua University and team members worked together and used a wide range of data ( country-wide, city-based) of 130 cities across China, and multi-level feature variables (e.g., socio-economic factors, natural conditions, and internal conditions) to establish a machine learning model MSW generation multi-cities with high accuracy. Their work analyzed and explored the waste management patterns of two typical large cities (Beijing and Shenzhen) in China. This study, titled “Development of a Multi-City Machine Learning Model for Municipal Solid Waste Generation Prediction,” is published online at Frontiers of environmental science and engineering.

In this study, a database of MSW generation and feature variables covering 130 cities across China was constructed. Based on the database, an advanced machine learning algorithm (GBRT) was adopted to build the waste generation prediction model (WGMod). In the model development process, the main factors influencing the generation of MSW were identified through weighted analysis. The major influencers selected were annual precipitation, population density, and annual mean temperature with weights of 13%, 11%, and 10%, respectively.

The WGMod showed good performance with R2=0.939. The model’s prediction on DSM generation in Beijing and Shenzhen indicates that waste generation in Beijing would gradually increase over the next 3-5 years, while that in Shenzhen would increase rapidly over the next 3 years. The difference between the two is mainly due to different trends in population growth.

This study established a database of MSW generation and feature variables with 1,012 datasets covering 130 cities across China. The developed WGMod works reasonably well and is very suitable for predicting the generation of MSW in China. This study provided scientific methods and basic data for the development of a multi-city model for the generation of MSW.


City’s digital twins help train deep learning models to separate building facades


More information:
Wenjing Lu et al, Development of a Multi-City Machine Learning Model for Municipal Solid Waste Generation Prediction, Frontiers of environmental science and engineering (2022). DOI: 10.1007/s11783-022-1551-6

Provided by Higher Education Press

Quote: Forecasting Municipal Solid Waste Generation Using a Multi-City Machine Learning Model (2022, September 26) Retrieved September 26, 2022 from https://techxplore.com/news/2022-09- municipal-solid-machine-multi-city. html

This document is subject to copyright. Except for fair use for purposes of private study or research, no part may be reproduced without written permission. The content is provided for information only.

Similar Posts

Leave a Reply

Your email address will not be published.