Machine-learning approaches in geo-environmental engineering: Exploring smart solid waste management.

Area of Science:

• Environmental Science
• Data Science
• Waste Management Engineering

Background:

• Domestic waste (DW) constitutes a significant portion of municipal solid waste (MSW), posing technical and financial challenges for municipalities.
• Accurate estimation, prediction, and characterization of DW are essential for developing sustainable MSW management strategies.
• Current methods face challenges in precisely forecasting DW generation.

Purpose of the Study:

• To estimate and predict domestic waste (DW) quantities using various machine learning (ML) algorithms.
• To evaluate the efficacy of different ML models in forecasting DW generation.
• To provide insights for developing smart waste management systems.

Main Methods:

• Employed multiple machine learning algorithms including linear regression, regression trees, Gaussian process regression, support vector machine, and autoregressive integrated moving average (ARIMA) for time series analysis.
• Conducted two case studies: one utilizing historical DW data (2010-2021) from Saudi and Bahrain authorities, and another tracking a family's waste generation for one month.
• Validated and tested models using metrics such as residuals, mean square error, root mean square error, and R²-Score.

Main Results:

• Biodegradable and non-biodegradable waste generated by the family ranged from 1.7-7.9 kg and 0.0-2.0 kg, respectively.
• Promising outcomes were achieved through judicious selection of input predictors and time series analysis.
• Model performance, indicated by R²-Scores, ranged from 0.67-0.85 for training and testing datasets across various predicted waste quantities.

Conclusions:

• Machine learning algorithms demonstrate significant potential for accurate domestic waste (DW) prediction.
• The study highlights the feasibility of using ML for developing smart waste management engineering systems.
• Effective DW prediction can mitigate the environmental, economic, and societal impacts associated with waste generation.