This analysis compares KNN, Random Forest, and MLR for predicting scope-1 CO₂ emissions, suggesting enhanced models may support Indonesia's emissions reduction goals.
The cement industry is one of the biggest contributors to the Indonesian total GDP. However, it also generates immense amounts of Green House Gas (GHG) emissions. This study is aimed at developing and comparing three predictive models, namely the K-Nearest Neighbor (KNN), Random Forest, and Multivariable Linear Regression (MLR) to create Scope-1 CO₂ emissions projection from Indonesian cement manufacturers. Using five years of historical data from cement plants across Indonesia, this study has identified key emission factors and the most suitable machine learning model to predict Scope-1 CO₂ emissions. Each model was evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R2 to assess their accuracy. Furthermore, this study also proposed an integration of these machine learning models into conventional emission calculation methods to provide a robust emissions projection framework that supports the Indonesian Net Zero goal.
No takes yet. Share an insight, caveat, or question.
Kaburuan et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: