01089385

Component

https://doi.org/10.3141/2064-10

http://scholar.google.com...Lint&publication_year=2008

Short-term prediction of travel time is a central topic in contemporary intelligent transportation system (ITS) research and practice. Given the vast number of options, selecting the most reliable and accurate prediction model for one particular scientific or commercial application is far from a trivial task. One possible way to address this problem is to develop a generic framework that can automatically combine multiple models running in parallel. Existing combination frameworks use the error in the previous time steps. However, this method is not feasible in online applications because travel times are available only after they are realized; it implies that errors on previous predictions are unknown. A Bayesian combination framework is proposed instead. The method assesses whether a model is likely to produce good results from the current inputs given the data with which it was calibrated. A powerful feature of this method is that it automatically balances a good model fit with model complexity. With the use of two simple linear regression models as a showcase, this Bayesian combination is shown to improve prediction accuracy for real-time applications such as dynamic traffic management, but the method is sensitive in the event that all models are biased in a similar way. It is therefore recommended to increase the number and the diversity of the prediction models to be combined.

01114862

English

pp 73-80

2008

Transportation Research Record: Journal of the Transportation Research Board

9780309113366

Print

Figures (5) ; References (17) ; Tables (1)

Advanced traffic management systems; Bayes' theorem; Mathematical models; Mathematical prediction; Regression analysis; Travel time

Data and Information Technology; Highways; Operations and Traffic Management; I73: Traffic Control

TRIS, TRB

Jan 29 2008 2:51PM

• Determining London Bus Stop Locations by Means of an Automatic Vehicle Location System
• Enhancing In-Car Navigation Systems with Personal Experience
• Evaluation and Selection of Adaptive Traffic Control Strategies on Transportation Networks: Decision Support Tool Based on Geographic Information System
• Forecasting Travel Supply and Demand by Modeling Multiscale Urban Environments
• Information Technology as a Critical Element of the Transportation Infrastructure: Critical Analysis in Modern Transportation Agencies
• Reducing Burden and Sample Sizes in Multiday Household Travel Surveys
• Research to Production: Lessons Learned from an Innovative Information System Development Approach
• Scenario-Driven Computer-Based Regional Incident Management Training
• Transformation of Transportation Data Models from Unified Modeling Language to Web Ontology Language
• Use of Global Positioning System to Identify Crash Locations in South Carolina