1. Prediction of traffic congestion based on time series dataset number of vehicles using neural network algorithm.
- Author
-
Yunanto, Prasetyo Wibowo, Gernowo, Rahmat, and Nurhayati, Oky Dwi
- Subjects
- *
TRAFFIC congestion , *TRAFFIC density , *TRAFFIC patterns , *CONGESTION pricing , *CYCLING training - Abstract
The increasing number of vehicles that was not accompanied by an adequate road infrastructure readiness causes traffic jams. Traffic jams often occur repeatedly every day, especially at certain times, such as when people are going to and from work. The same traffic jams can also occur at the beginning or the end of the week; this usually repeats every week. As a result, if the congestion dataset is known, the repeating traffic congestion for daily and weekly congestion can be anticipated. In this study, traffic congestion predictions are modeled using the Neural Network Algorithm based on traffic congestion data collected within 24 hours for one to two weeks. The parameters used in optimizing Neural Network performance are learning rate, momentum, and epochs (training cycles). Based on these experiments, the Neural Network Algorithm successfully modeled traffic density patterns which are recurring congestion patterns, with quite good results. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF