Data-Driven Household Waste Management Research

Urban waste management is a growing challenge in modern cities today. Low-income municipalities like

Quinta Normal face strong financial and operational constraints. Currently, reactive planning without

historical data leads to increased operational costs and high service variability. Highly granular data

collected weekly enables municipalities to implement precise predictive planning models. Advanced machine learning

models like Random Forest support accurate weekly demand forecasting. The result is a more efficient,

inclusive, and sustainable waste collection system.