Prediction of municipal waste generation using multi-expression programming for circular economy: a data-driven...

Area of Science:

• Environmental Science
• Computer Science
• Sustainable Development

Background:

• Rising municipal waste generation (MWG) presents a significant obstacle to sustainable development.
• Effective waste management strategies require accurate predictive models, especially within the circular economy framework.

Purpose of the Study:

• To develop and validate a novel predictive model for municipal waste generation (MWG) using multi-expression programming (MEP).
• To compare the performance of the MEP model against artificial neural network (ANN), random forest (RF), and multiple linear regression (MLR) models.
• To identify key socioeconomic and environmental factors influencing MWG through sensitivity analysis.

Main Methods:

• Development of a predictive model using multi-expression programming (MEP) based on historical socioeconomic and environmental data.
• Validation of the MEP model through comparative analysis with ANN, RF, and MLR models using various evaluation metrics.
• Conducting parametric and sensitivity analyses to assess the MEP model's performance and the influence of input variables.

Main Results:

• The MEP model achieved a higher coefficient of determination (R² = 0.977) compared to ANN (R² = 0.974), MLR (R² = 0.964), and RF (R² = 0.957) for predicting MWG.
• Sensitivity analysis identified the relative importance of input variables in the MEP model.
• The study formulated a new mathematical model linking socioeconomic and environmental factors to MWG.

Conclusions:

• The MEP model demonstrates superior accuracy and performance for predicting municipal waste generation (MWG).
• The developed model provides a valuable tool for waste management authorities to optimize infrastructure and policies for a circular economy.
• This research contributes a novel MEP application and a mathematical model for informed waste management and sustainable development.