<![CDATA[Traffic is one location where people spend a lot of time. Currently with the rapid growth of vehicles makes conditions in Indonesia is getting crowded every day. One way to solve this problem is with traffic lights. However, the current traffic light performance is considered less than optimal. Therefore required a system that can determine the time, so the time at the traffic light can be more dynamic based on traffic conditions. This research uses backpropagation method to determine the duration of traffic lights based on queue lenght of vehicle. The result of the trained data test obtained is the Linear function with a = 3, the optimal iteration obtained at iteration 10, the optimal number of nodes in the hidden layer is 2, and the optimal value of learning rate is 0,02. The evaluation result when processing the test data using the optimal activation function, the optimal number of iterations, the optimal number of nodes in the hidden layer, and the optimal learning rate yields RMSE value of 0,0888978841028.]]>
Copyrights © 2018
