Study of Traffic Flow Forecasting Based on Genetic Neural Network
Managers of transportation systems now have access to large amounts of real-time status data and a variety of methods and techniques have been developed to forecast traffic flow. This paper, presented at the 6th International Conference on Intelligent Systems Design and Applications (Jinan, China, 2006) presents a study that used traffic flow forecasting based on a genetic neural network. Traffic flow forecasting models based on neural networks have been applied widely in intelligent transportation systems (ITS) with high forecasting accuracy and self-learning ability. However, there are problems in neural networks, such as the difficulty of designing optimal structures and weak global searching capabilities, which limit its applications. This paper uses the genetic algorithm, which has a powerful global exploration capability, to solve the problems of tuning both network structure and parameters of a feedforward neural network. After the authors introduce the genetic neural network algorithm in detail, they demonstrate how this approach can be applied to traffic flow forecasting.
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/isbn/0769525288
-
Corporate Authors:
IEEE Computer Society
1730 Massachusetts Avenue, NW
Washington, DC United States 20036 -
Authors:
- Ji, Tao
- Pang, Qingle
- Liu, Xinyun
-
Conference:
- 2006 6th International Conference on Intelligent Systems Design and Applications
- Location: Jinan , China
- Date: 2006-10-16 to 2006-10-18
- Publication Date: 2007
Language
- English
Media Info
- Media Type: Print
- Features: References;
Subject/Index Terms
- TRT Terms: Algorithms; Forecasting; Genetic algorithms; Intelligent transportation systems; Neural networks; Optimization; Real time information; Traffic flow; Traffic forecasting
- Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning;
Filing Info
- Accession Number: 01054236
- Record Type: Publication
- ISBN: 0769525288
- Files: TRIS
- Created Date: Jul 23 2007 6:13PM