<![CDATA[Congestion is one of the main problems for big cities in Indonesia. This is influenced by the growth number of vehicles in Indonesia that are not matched by road growth. So, congestion cannot be avoided. In addition to these factors, the static traffic system in Indonesia contributes as a cause of congestion. at a traffic junction, sometimes a segment must wait for waiting time, while in other segments that intersect have no queue. So in this research of traffic light optimization system was designed using the naive bayes classifier method which was simulated using the SUMO Simulator Software and will communicated with Arduino through serial communication. In this research the density or length of the queue and each section will be used as input system that will be processed with training datas. The output system is the duration of traffic light that adjust by density in each sections. From the tests carried out, the suitability of naive bayes classifier with training datas reaches 100%. The Arduino and SUMO Simulator suitability tests reached 100% and the traffic light system optimization test using the naive bayes classifier produces a percentage of 86,66% better than the traffic system without naive bayes classifier of 15 times the side carried out.]]>
Copyrights © 2020
