Technology Platform
Traffic Data Collection
  ----Incident Data
  ----Flow Data
  ----Traffic Probe
Data Fusion Engine
Traffic Modeling
Application
Patents

Facts About Probe Data

Based on a Study by General Motors in 2004-2005, to support a real-time traffic information system based on probe vehicles:

. at least 3% probe samples or 5 million probe vehicles are required to cover freeways in the US, while more than 5% or 8-10 million probe vehicles are required to cover major surface streets in the US ;
. for top 10 metropolitan areas in the US, at least 2.3 million probe vehicles are required to cover freeways and at least 3.8 million probe vehicles are required to cover major surface streets;
. for top 30 metropolitan areas in the US, at least 3 million probe vehicles are required to cover freeways and at least 6 million probe vehicles are required to cover major surface streets.

Sources:

    • Journal of Intelligent Transportation Systems, Volume 9, Number 1, January-March 2005
    • US Census 2000
    • US Bureau of Transportation Statistics
  • Based on a Study by the University of California at Berkeley in 2000, to support a real-time traffic information system based on probe vehicles in the San Francisco Bay Area, 5% samples or 300,000 probe vehicles are required to cover freeways.

Sources:

    • PATH, University of California at Berkeley, Working Paper, PaperUCB-ITS-PWP-2000-18, 2000.

TCI Probe Technology

To improve model accuracy, TrafficCast International is currently incorporating floating vehicle data (FVD) collected using GPS devices, Wi-Fi Hot Spot technology, and cellular probe technology. Having adopted the probe data technologies from the ADVANCE Project (the largest traffic probe demonstration project in the US from 1991-1996 with a price tag of $50M), the TrafficCast International's models have designed modules to accommodate the latest input from probe vehicles. In strategic partnership with China Mobile and TrafficCast China, TCI has deployed the largest cellular probe system in Shanghai (http://www.trafficcast.com/news_06.11.01.htm), which produces travel speed and travel time information in real-time by using cellular probe data and 10 million China Mobile subscribers in Shanghai.

Currently, due to its limited sample size and limited coverage in terms of time and roads, the GPS probe data mainly function as historical benchmark. As more GPS probes are deployed and integrated into the data stream in the next 3-10 years, GPS probes will start to function more as a real-time input for real-time and predictive traffic flow data. However, this process is lengthy and requires significant investment by the industry.

On the other hand, cellular probe data could become a viable alternative to provide both real-time and historical flow data in the next 1-3 years. Because the cost to deploy cellular probe is much less than GPS probe and the time to provide large coverage is also much shorter. As part of its historical flow database, TCI integrated GPS probe data generated by fleet companies. However, the impact of GPS probe data will be limited until larger samples are available in real-time.