<![CDATA[This study aims to classify package delivery duration based on the difference between shipping time and delivery completion time using the Decision Tree algorithm. The dataset used in this research was obtained from PT Idenative and consists of historical logistics data containing various shipment attributes such as number of items, destination location, delivery status, and time-related variables. Data preprocessing was conducted through cleaning, transformation, and categorization, where delivery duration was classified into four categories: Fast, Normal, Slow, and Very Slow. The classification model was developed using the Decision Tree algorithm due to its interpretability and ability to handle both categorical and numerical data. The dataset was divided into training and testing sets with an 80:20 ratio, and model performance was evaluated using confusion matrix, accuracy, precision, recall, and F1-score metrics. The results show that the model achieved an accuracy of 40.72%, with a macro precision of 0.62, recall of 0.35, and F1-score of 0.34, indicating moderate performance. The model faces challenges in distinguishing between similar classes, particularly Normal and Slow categories. Feature importance analysis reveals that the number of items and destination location are key factors influencing delivery duration. This study demonstrates that the Decision Tree algorithm can be applied to classify delivery duration in the logistics domain while providing interpretable insights for operational decision-making. However, further improvements are required, such as applying ensemble methods and data balancing techniques to enhance model performance.]]>
Copyrights © 2026
