The Sinocanadian

In response to the results recent Beijing Energy Network – Environmental Challenge Open competition, which is is summarized by event judge, Kunal Sinha (Ogilvy & Mather) here, I thought I would just make a few comments on where real-time traffic information comes from in China.

Real Time Traffic - Central Beijing. (Google Maps)

Use of real-time traffic information for implementing smart-routing through GPS systems would be great for reducing traffic jams, and the emissions and energy consumption associated with them.  However, real-time traffic information comes from a complex data monitoring, evaluation and reporting mechanism that is far from perfect.

At this time in China, most real-time traffic data (as well as digital maps) come from two big GPS data companies, Autonavi and Sennavi ( I think that’s the English name).  They sell that data to Google, Baidu, municipal governments and other companies.

According to my understanding, they get this data from a number of sources:

• Taxis with GPS/GPRS systems mounted inside – These companies pay the taxi companies for the data provided by these GPS/GPRS systems.  The taxis drive normally, and report back their location and speed to the system which analyses their travel pattern.  The data from each car individually is not reliable because you never know why a car is stopped.  So I think they only accept data from these cars when they have a fare (the meter is running, so the car is not likely to stop and wait around — correct me if I’m wrong).  Only when there are many cars reporting their travel can the system make accurate predictions about the traffic situation at any particular location.  So, if there are no or few of these taxis in a particular area, other methods need to be used to gather traffic data.
  • This is particularly important, and is the major reason that traffic data is only available for major roads, and not for small roads:  if there are no reporting vehicles in a particular place, there is no way to know if there is a traffic problem or not.

• Road sensors – road sensors can show when traffic is stopped or moving — but these are expensive and need to be calibrated from time to time.  Therefore, it is difficult to install them on smaller roads.

• Cameras – Like road sensors, cameras can detect when traffic is stopped or moving.  I wonder if this is a potential area of expanding data on smaller streets regarding traffic movement, but this also brings into question of privacy and national security of video data.

• Call-in reports – these call in reports, while up-to-date, cannot be relied on to always report traffic issues, and need to be verified.  They also require actual people to answer phones/inquiries, meaning increased labour cost.

After data is collected, it has to be processed, and judged for accuracy.  If data is suddenly unavailable, historical systems have to be ready in order to predict the likely traffic situation.  Finally, the data is broadcast by FM-radio signal across to GPS terminals across the city, or through the internet on portals such as google or Baidu maps.

Needless to say, this is a capital-intensive and heavily regulated service, and is not easy to get into as a new entrant.  Apparently, data is available on major roads in Beijing, Shanghai, Shenzhen, Guangzhou, Chengdu and several other major centers.

In this way, it seems that Beijing in particular has challenges for providing a large number of alternative routes around traffic jams based on real-time traffic information. First, most providers of real-time traffic data are already on major roads.  Data about smaller roads (as possible alternative routes) will be extremely limited and at best, unreliable.  Secondly, due to Beijing’s ring-road / large city block structure, there are relatively few alternative routes to take once one has already set off, particularly on cross-city trips.

I guess there must be other challenges to solve here, but these are some of the key ones.  I hope somebody can figure them out!

By the way, iCET is working with Autonavi to try and offer a GPS-based GHG emission handheld application (it’s only for Symbian phone at the moment, and only in Chinese).  The first beta edition can’t do it yet — but it is already in Autonavi’s Minimap platform.  After a bit more work, we should have real GPS-based transport emission data in a handheld application based on mode of travel, including specific vehicle emissions.