<![CDATA[The growth of vehicle volume in Malang City presents challenges in the form of increased accident risks, especially if the technical condition of the vehicles does not meet standards. As a preventive measure, the Motor Vehicle Feasibility Testing (KIR Test) is conducted to ensure that vehicles comply with safety standards. However, manual assessments in this process are prone to human error, necessitating a more efficient and accurate system. This study implements the Random Forest method to classify the eligibility status of motor vehicles, focusing on two main categories: public and private vehicles. This implementation is expected to improve the efficiency and accuracy of the KIR test process. Among the data split ratios tested, a 60% training data and 40% test data ratio yielded the best results with an accuracy of 86.94% and an OOB error rate of 13.03%, indicating the model's error rate on data not used during training. These results indicate that the Random Forest method effectively identifies the eligibility status of motor vehicles with an optimal data configuration.]]>
Copyrights © 2025
