New data sources for solving traffic problem
While it is still a prototype system, DynaMIT can already handle some
of the large-scale real-world networks. Of course there are still a
few issues to be resolved before systems like DynaMIT could become
commercially viable products. For example, so far, lacking reliable data sources (especially real-time sources) remains one of the major problems.
Recently I have been working on projects that use new data sources to predict traffic conditions. Things we are considering include many different types of mobile devices -- cell phone, Wi-Fi devices, RFID transponders, GPS-equipped vehicles (e.g., taxi fleet), electronic toll collections systems, etc. New technologies such as the one (recently used by Google) described in
this article on Technology Review, might eventually lead to more accessible data (in terms of the cost of collecting). DynaMIT can be extended to "fuse" these data together to forecast congestion in the immediate future and generate travel guidance.
With the rapid development of wireless telecommunication and mobile computing technology, I believe in a few years mobile devices by themselves would probably generate enough data to fill in the gap between what is available today and what is needed for the real-time analysis and prediction of traffic congestion. In the near future some company might be able to make use of the ubiquitous data and provide to travelers various anti-congestion services that are much superior to what we can do today.