Loading...
Please wait, while we are loading the content...
Similar Documents
Short-term travel time prediction using a neural network
| Content Provider | Semantic Scholar |
|---|---|
| Author | Huisken, Gio Berkum, Eric C. Van |
| Copyright Year | 2002 |
| Abstract | The growth in car mobility has lead to more uncertainty in travel times. As a result cardrivers have an increasing demand for information on these travel times (see e.g. Huisken and Van Berkum, 2000). Travel times can be measured using automated vehicle identification (AVI) techniques (floating car data, automated license plate recognition, etc.). However, these techniques are rarely used since they need large investments in roadside equipment. Currently travel times are estimated using data from inductive loop detectors. Since loop detectors yield spot measurements of flow and speed, travel times can only be estimated, not actually measured. Furthermore, when cardrivers are provided with information on travel times, these travel times should ideally be the times that they will encounter. Therefore we need to predict travel times, based on previous measurements. Currently two methods are being used, i.e. Static Travel Time Estimations (STTE) and Dynamic Travel Time Estimations (DTTE) (see e.g. Hounsell and Ishtiaq, 1997; Van Arem et al, 1997; and Zee, 2001). This research proposes a new travel time prediction method using an Artificial Neural Network (ANN). The three methods STTE, DTTE, and ANN methods were applied on the A13 motorway from The Hague to Rotterdam and their performance was compared. |
| Starting Page | 93 |
| Ending Page | 98 |
| Page Count | 6 |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.iasi.cnr.it/ewgt/13conference/17_huisken.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |