A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prediction of the Impact of Typhoons on Transportation Networks with Support Vector Regression
The ability to predict the impact of typhoons on transportation infrastructure is important as it can help to avoid serious delays and dangers when roads are closed due to such events. This research uses support vector regression (SVR) to predict the impact of typhoons on transportation infrastructure. It first integrates and examines the infrastructure and precipitation data from different authorities. An SVR model is constructed to solve a nonlinear prediction problem for small size data. The SVR model is calibrated and validated by a heuristic process. The calibrated and validated results are then applied to predict closed roads in a real network through a simulation assignment model. Several traffic management strategies are developed to reduce the negative impacts of typhoons. The results show that the mean absolute percentage error (MAPE) of SVR prediction is 9.7%. The impact of typhoons on transportation networks can thus be predicted and simulated based on the calibrated SVR model, and appropriate strategies can then be developed in order to reduce both delays and risks.
Prediction of the Impact of Typhoons on Transportation Networks with Support Vector Regression
The ability to predict the impact of typhoons on transportation infrastructure is important as it can help to avoid serious delays and dangers when roads are closed due to such events. This research uses support vector regression (SVR) to predict the impact of typhoons on transportation infrastructure. It first integrates and examines the infrastructure and precipitation data from different authorities. An SVR model is constructed to solve a nonlinear prediction problem for small size data. The SVR model is calibrated and validated by a heuristic process. The calibrated and validated results are then applied to predict closed roads in a real network through a simulation assignment model. Several traffic management strategies are developed to reduce the negative impacts of typhoons. The results show that the mean absolute percentage error (MAPE) of SVR prediction is 9.7%. The impact of typhoons on transportation networks can thus be predicted and simulated based on the calibrated SVR model, and appropriate strategies can then be developed in order to reduce both delays and risks.
Prediction of the Impact of Typhoons on Transportation Networks with Support Vector Regression
Ta-Yin Hu (author) / Wei-Ming Ho
2015
Article (Journal)
English
Prediction of the Impact of Typhoons on Transportation Networks with Support Vector Regression
| Online Contents | 2015
| British Library Online Contents | 2007
Travel time prediction with support vector regression
| IEEE | 2003
Travel Time Prediction with Support Vector Regression
| British Library Conference Proceedings | 2003