Classification and Prediction of Traffic Flow Based on Real Data Using Neural Networks
Journal Title: Archives of Transport - Year 2012, Vol 24, Issue 4
Abstract
This paper presents a method of classification of time series of traffic flow, on the section of the main road leading into the city of Gliwice. Video detectors recorded traffic volume data was used, covering the period of one year in 5-minute intervals – from June 2011 to May 2012. In order to classify the data a statistical analysis was performed, which resulted in the proposition of splitting the daily time series into four classes. The series were smoothed to obtain hourly flow rates. The classification was performed using neural networks with different structures and using a variable number of input data. The purpose of classification is the prediction of traffic flow rates in the afternoon basing on the morning traffic and the assessment of daily traffic volumes for a particular day of the week. The results can be utilized by intelligent urban traffic management systems.
Authors and Affiliations
Teresa Pamuła
Keywords
Related Articles
- EP ID EP70067
- DOI 10.2478/v10174-012-0032-2
- Views 156
- Downloads 0
Mechanisms for increasing of transportation efficiency using joint service of logistics systems
More and more attention become to Transport Company’s functioning efficiency due to growing of goods’ nomenclature and specific requirements for their service. Existing scientific and practical approaches to managing of...
A study framework for assessing the performance of the urban freight transport based on PLS approach
The urban freight transport (UFT) is an important component of the urban logistics. It represents a driving force for the economic dynamics and attractiveness of a city. It can be treated as all movements of goods necess...
Decision-making model for supporting supply chain efficiency evaluation
This paper presents some approach for the formulation of the decision-making model in supporting the assessment of supply chain efficiency. Were presented factors affecting the efficiency of the functioning of individual...
Non-parametric machine learning methods for evaluating the effects of traffic accident duration on freeways
Traffic accidents usually cause congestion and increase travel-times. The cost of extra travel time and fuel consumption due to congestion is huge. Traffic operators and drivers expect an accurately forecasted accident d...
Reliability and availability assessment of diesel locomotive using Fault Tree Analysis
The article presents an application of a method based on fault tree analysis and the Monte Carlo simulation in the assessment of reliability and availability of the rail means of transport. The primary target of the pres...