Forecasting of municipal solid waste generation in China based on an optimized grey multiple regression model

The massive generation of municipal solid waste (MSW) has become an essential social problem that not only damages the ecological environment but also affects human health. To effectively manage MSW, it is necessary to forecast waste generation accurately. In this study, a grey multiple non-linear regression (GMNLR) model is developed to achieve the effective forecasting of MSW generation in China. Using grey relational analysis (GRA) to rank the influential factors of MSW generation, it is found that urban road area, residential consumption level, and total population are the main factors. Then, these factors are used as the input variables of the model to forecast MSW generation. Meanwhile, four performance indicators with adjusted R-2 (R-adj(2)), absolute percentage error (APE), mean absolute percentage error (MAPE), and root mean square error (RMSE) are used to evaluate the performance of these models. The results demonstrate that the GMNLR model has a highest prediction accuracy among the four models. According to the forecast results, China's MSW generation will reach 332.41 million tons in 2025, with an annual growth rate of 8.28%. The combined model proposed in this paper is helpful for the government in policies and regulations making for MSW management.

Automated summary

In this study, a grey multiple non-linear regression (GMNLR) model was developed to accurately forecast municipal solid waste (MSW) generation in China. The influential factors of MSW generation were ranked using grey relational analysis (GRA), and it was found that urban road area, residential consumption level, and total population were the main factors. The GMNLR model showed the highest prediction accuracy among the four models evaluated.