A hybrid machine learning-mathematical programming optimization approach for municipal solid waste management during...

Area of Science:

• Environmental Science
• Operations Research
• Computer Science

Background:

• The COVID-19 pandemic significantly impacted municipal solid waste management systems globally.
• Effective waste management strategies are crucial for public health and environmental protection during epidemics.

Purpose of the Study:

• To develop and evaluate a mathematical optimization and machine learning strategy for optimal municipal solid waste management during the COVID-19 epidemic.
• To integrate optimization models for supply chain determination with machine learning for parameter estimation and future projections.

Main Methods:

• Developed an optimization model using the General Algebraic Modeling System (GAMS) to determine the optimal waste management supply chain.
• Utilized machine learning prediction models in Python to estimate time-dependent parameters and generate future waste management projections.
• Applied the strategy to a case study in New York City, using extensive socioeconomic data to train the machine learning model.

Main Results:

• Successfully predicted municipal solid waste collection trends over time based on socioeconomic data.
• Identified significant trade-offs between economic objectives (profit) and environmental objectives (waste reduction in landfills).
• Demonstrated the strategy's utility for planning and decision-making in potential future pandemic scenarios.

Conclusions:

• The integrated optimization and machine learning approach provides a robust framework for adaptive municipal solid waste management during pandemics.
• The findings highlight the complex interplay between economic and environmental factors in waste management, offering valuable insights for policymakers.
• The study provides a data-driven tool for optimizing waste management strategies in response to public health crises.