Aisyah Larasati, Wahyu Sakti Gunawan Irianto, Apif Miftahul Hajji, Nafi Kareem
Industrial heavy equipment machines, such as backhoes used in various construction and mining projects, are one of the main sources of NOx (nitrogen oxides) gas emissions. This study aims to build an optimized Random Forest Regression (RFR) model using a grid search optimization to predict NOx emissions from backhoe engines. The data used in this study is secondary data from 37528 backhoe engine operational data, including variables such as backhoe engine type, engine power, MAP (manifold absolute pressure), engine speed, engine age, engine tier technology type, and engine temperature. The data was processed through several stages, including data cleaning, Pearson correlation test, Z-score standardization, and K-Fold Cross-Validation. The model performance is evaluated based on RMSE (root mean square error), MAPE (mean absolute percentage error), and R2. Based on the test, the best RFR model after grid search optimization obtains an RMSE value of 0.0065, a MAPE of 14%, and an R2 of 0.8894. The top three factors that affect the NOx gas emission from the highest to the lowest are RPM (revolutions per minute), MAP, and engine temperature. Based on this result, the operation of the backhoe engine should be maintained at the optimal condition of RPM, MAP, and engine temperature in order to minimize NOx gas emissions. © 2025 IEEE.
Universitas Negeri Malang, Dept. of Mechanical and Industrial Engineering, Malang, Indonesia; Universitas Negeri Malang, Dept. Electrical Engineering and Informatics, Malang, Indonesia; Universitas Negeri Malang, Dept. Building Maintenance Engineering and Technology, Malang, Indonesia
