<![CDATA[Traffic congestion is a major problem in Palembang City due to the significant growth in the number of vehicles. This study aims to develop an artificial intelligence-based system for detecting vehicle density and predicting optimal routes. Vehicle number detection is carried out using the YOLOv11 method based on CCTV data at 15 intersections in Palembang City, with training results showing an accuracy of 92%, F-1 Score of 82% and mAP@0.5 of 86.7%. In the validation and testing stages, this model achieved an accuracy of 90%, and mAP@0.5 of 81.7%. The detection data was then analyzed using the Random Forest (RF) algorithm to classify traffic conditions with a dataset of 769 rows of data, achieving an accuracy of 98.26%. Furthermore, the Bee Colony Optimization (BCO) algorithm was used to determine the fastest route by taking into account the distance traveled and the level of congestion. The results of the study show that the combination of the YOLOv11, RF, and BCO methods is able to produce an effective system in providing optimal route recommendations and helping to significantly reduce congestion. This system is expected to be a practical solution for city traffic management in the future.]]>
Copyrights © 2025
