Can technology save the environment?

I have recently purchased a new vehicle with a litres per hundred kilometres active display. This monitors fuel mileage as you drive. What I noticed was both disturbing and a potential place for vast improvement in terms of how technology can help the environment. I hereby give this idea up freely to anyone wishing to pursue it further - this IP is potentially too helpful to have anyone patent it.

I noticed that my vehicle on flat ground cruised along at around 5l/100km (around 55 MPG) through green lights on relatively flat ground. If I had to stop and re-start, the gauge went to 99l/100 km (about 3 MPG) to re-accelerate. After some further testing and calculations, I verified the hypothesis that starting and stopping a vehicle can account for around 80% + of the fuel used vs. flowing traffic. I drove from Whistler to North Vancouver (around 111 km) and used the same amount of gas I used to go the last 15 km through Vancouver.

What does this mean?

Other than validating a few basic laws of physics (conservation of angular momentum, bodies in motion tend to stay in motion, etc.), this means that our governments and almost the entire industry trying to focus on getting better mileage for vehicles is basically inconsequential in comparison to better active/adaptive traffic systems. In fact, other than the move to hybrids (which the jury is still out on in terms of NET energy savings), even if you bettered the mileage of every vehicle by 50%, it would mean a 5% gain compared to a better traffic flow which could deliver potentially up to 50% better return.

While it would not be possible to have green lights for everyone, there are certainly situations I have encountered where a high volume road is sometimes held up for the sake of one or two cars to utilize the right of way of an intersection.

An interim step could be to replace traffic lights that meet certain thresholds with roundabouts to keep traffic from having to completely stop in all directions. Hybrid cars are also good as they rely on capturing brake energy and being able to store it as electricity then use the electricity to re-accelerate the car. IMO - these are good intermediate steps but the holy grail is in a different place.

Opportunity:

What if every vehicle had some form of transponder that could communicate its current MPG to sensors in the road as well as the road being able to sense how many cars are traveling in each direction. A grid of sensors could be used to feed an array of every vehicle on the road into a central planning computer which could optimize the traffic flow (within parameters) for the best fuel efficiency.

I am an environmentalist but I am also a pragmatist. While some advocate that everyone should ride bikes and take public transit, I just don’t see the auto industry silently going away anytime soon. This means an immediate focus on improving hydrocarbon efficiency is probably still warranted IMO.

This is obviously a totally unproven hypothesis however there is sufficient evidence to warrant a test simulation IMO. Even if this proved to be consequential, such a system would take decades to implement.

There are other benefits as well. Now imagine that the transponder could be wired into your electronic dashboard and warn you of events on the road ahead (example - "unexpected hard braking - 100 meters ahead")? You could even have audio communications coming in to the car as instructions to the driver such as "maintain 45 KPH and you will make the next 4 stop lights. If you go faster, you will have to brake and re-accelerate".

This is not some wild pipe dream. All the technology exists today.

Question:

Does this have merit?
Would it be worthwhile conducting some further studies?
Is anyone interested in discussing something like this at FooCamp (or somewhere else) this year (assuming I will be going again)?