Landfill area estimation based on integrated waste disposal options and solid waste forecasting using modified ANFIS...

Area of Science:

• Environmental Science
• Waste Management Engineering
• Artificial Intelligence in Environmental Applications

Background:

• Accurate solid waste data collection is a significant challenge in developing nations.
• Solid waste generation is influenced by dynamic economic, demographic, and social factors.
• Minimizing land requirements for waste disposal is critical for sustainable urban planning.

Purpose of the Study:

• To develop a robust solid waste prediction model for Malaysia.
• To identify optimal input factors for solid waste forecasting.
• To reduce land requirements for solid waste disposal in alignment with Malaysian environmental goals.

Main Methods:

• Utilized a modified adaptive neural fuzzy inference system (MANFIS) for solid waste prediction.
• Integrated solid waste forecasting, waste composition data, and the Malaysian vision for waste disposal.
• Evaluated model performance using root mean square error (RMSE) and coefficient of determination (R²).

Main Results:

• The MANFIS model achieved high accuracy with RMSE for training at 0.2678 and for testing at 3.9860.
• The coefficient of determination (R²) reached 0.99, indicating excellent model fit.
• Implementation of the Malaysian vision, guided by the prediction model, can reduce land requirements by up to 43%.

Conclusions:

• The developed MANFIS model provides a reliable method for solid waste prediction.
• Accurate forecasting enables significant reductions in land needed for waste disposal.
• This approach supports sustainable waste management and aligns with national environmental objectives.