Modelling municipal waste generation using support vector machine, artificial neural network and deep learning

Articles in Press

Document Type : Research Article

Authors

Faculty of Civil, Water and Environmental Engineering, Shahid Beheshti University, Tehran, Iran

Abstract

The purpose of this research is to investigate and compare the performance of intelligent models in quantitative modeling of urban waste. First, the factors affecting waste production, including geographical, social, meteorological, cultural, and economic information were collected monthly and seasonally. Then, quantitative modeling of urban waste in Tehran was done using intelligent models of artificial neural network, support vector machine and deep learning. and the results and errors obtained from them have been investigated. According to the modeling done, it was concluded; The regression model and the artificial neural network have the lowest R2 and the highest RMSE and MAE and do not perform accurate modeling. Based on the criteria and errors obtained, it was concluded that deep learning, support vector machine model, artificial neural network, and regression in the last rank have worked in accurate modeling both in the monthly period and in the seasonal period. The support vector machine model and the deep learning model have the least errors among the tested models both in the seasonal period and in the monthly period. In the monthly modeling, the figures observed are closer to the figures predicted by the deep learning model than other models and are more consistent, in addition, the deep learning model in seasonal modeling is more accurate than the other tested models. It should be noted that the deep learning algorithm in modeling A season of monthly modeling has been more accurate;

Keywords

Main Subjects

Amirkabir Journal of Civil Engineering

Articles in Press, Accepted Manuscript
Available Online from 14 January 2025

Files

Share

How to cite

Statistics