Applying the Efficiency Analysis Tree Method for Enhanced Eco-Efficiency in Municipal Solid Waste Management: A Case Study of Chilean Municipalities

Enhancing the eco-efficiency of municipal solid waste (MSW) services is pivotal for the shift toward a circular economy. Although the Data Envelopment Analysis (DEA) method is widely used, it is susceptible to overfitting, potentially distorting eco-efficiency assessments. This study applies the efficiency analysis tree (EAT) method, which synergizes machine learning and linear programming, offering a more reliable framework for eco-efficiency evaluation in the MSW sector. This innovative approach provides deeper insights into the optimal levels of operational costs and unsorted waste. The research encompasses a case study of 98 Chilean municipalities from 2015 to 2019, uncovering significant disparities in optimal operational expenses and unsorted waste quantities, which underscores the necessity for customized waste management approaches. The average eco-efficiency scores for 2015–2019 range between 0.561 and 0.566. This means that assessed municipalities can reduce unsorted waste by amounts ranging from 1,632,409 tons/year (2016) to 1,822,663 tons/year (2018). Potential economic savings estimated are 105,973 USD/year (2019), which represents 44% of the total MSW management costs. Additionally, the investigation into the effects of external factors on eco-efficiency furnishes nuanced perspectives that can guide policymakers and municipal authorities in developing effective, context-specific waste management strategies. Beyond refining eco-efficiency evaluations, this study contributes to more informed decision-making processes, aiding the progression toward sustainable waste management practices.