| Transportation Help Desk | |
| Traffic Safety Evaluations | |
| Library Services | |
| Video Library | |
| Going... Going... Gone | |
| Ask an Expert | |
| Tech Transfer Newsletters | |
| Publications | |
| Free ITS Training | |
| Join Our Mailing List | |
| Regional Planning Help | |
| Title: | The I-880 Field Experiment: Effectiveness of Incident Detection Using Cellular Phones |
| Authors: | Alexander Skabardonis, Ted Chira-Chavala, Daniel Rydzewski |
| Date: | 1998 |
| Call No: | UCB-ITS-PRR-98-1 |
ProblemCellular phones can be an inexpensive alternative to other incident-detection technologies, involving little capital or infrastructure costs. However, calls by motorists can vary greatly in detail and quality of information, and in the timeliness of reports. How well do they perform compared to other methods? MethodWe compared incidents reported by cell phones along a nine-mile section of I-880 in the Bay Area to data obtained from field observation and other detection sources, creating a database of 264 incidents (accidents and lane blockages) and 1429 "other events" (stalls and events other than accidents that did not block lanes). We measured the quality and adequacy of cell phone data in terms of the incident detection rate, false alarm rate, timeliness, and available details such as location, type, and severity of incident, and type and number of vehicles involved. FindingsCell phone callers detect 38% of incidents and 1% of other events. 7% of calls about incidents were false alarms, and 32% of calls about other events were false alarms. Cell phone users report an additional 35% of incidents that are not witnessed by the CHP; the CHP and cell phone users together detect 60% of incidents. Relative to other sources, cell phones have the highest detection rate and are the fastest detection source. However, they have a high false alarm rate (perhaps due to the fact that drivers can't tell whether a car by the side of the road is stalled or just temporarily stopped), and a low detection rate for other events (perhaps because drivers don't pay attention to events unless they present an immediate hazard). Also, callers provide limited information on the severity of incidents, and the calls need verification, which contributes to the fact that incidents detected by cell phone users take longer to clear than CHP-detected incidents. Cellular phones are not enough by themselves to detect incidents and other events, but they could be more effective if the CHP changed verification and response practices |
|
In the box below, type a word or phrase:
(Examples:
Use your browser's "Back" button to return to listing