Forecasting the electronic waste quantity with a decomposition-ensemble approach.

Area of Science:

• Environmental Science
• Data Science
• Waste Management

Background:

• Waste electrical and electronic equipment (WEEE), or e-waste, represents the fastest-growing category of hazardous solid waste globally.
• Effective e-waste disposal and management strategies depend on accurate quantity predictions.
• Existing forecasting models may not fully capture the complex trends in e-waste generation.

Purpose of the Study:

• To propose a novel hybrid forecasting methodology for predicting e-waste quantities.
• To enhance the accuracy and reliability of e-waste amount predictions.
• To support the development of effective e-waste recycling and circular economy initiatives.

Main Methods:

• A decomposition-ensemble-based hybrid forecasting methodology integrating Variational Mode Decomposition (VMD), Exponential Smoothing Model (ESM), and Grey Modeling (GM) was developed (VMD-ESM-GM).
• The VMD-ESM-GM model decomposes complex time-series data into simpler components for improved analysis.
• The methodology was validated using e-waste data from Washington State, US, and the UK Environment Agency.

Main Results:

• The VMD-ESM-GM methodology demonstrated superior performance compared to benchmark models in predicting e-waste quantities.
• The model accurately captured and predicted future fluctuation trends in e-waste generation.
• The decomposition-ensemble approach proved effective in handling the complexities of e-waste data.

Conclusions:

• The proposed VMD-ESM-GM hybrid model is a suitable and effective tool for e-waste quantity prediction.
• Accurate e-waste forecasting can significantly aid decision-making for recycling and circular economy planning.
• This methodology offers a robust approach for managing the growing challenge of global e-waste.