Caution: I'm not an engineer, so don't look for a lot of science or detail behind this.
Potholes suck, particularly in areas hit especially hard by our current but soon to be past winter. I've already replaced one wheel as a result of hitting a particularly heinous pothole on the highway, one that took out more than a dozen cars a couple Sundays ago.
From my basic understanding, potholes are caused by a failure of the road surface that allows water to leak between layers of asphalt. The water freezes in the winter, expanding and cracking the surface, which then crumbles as vehicles pass over it. It is particularly bad in places where there are multiple layers of asphalt. Here in Connecticut for example, they grind down the old roadway to give the new asphalt something to bite to, but water still finds a way of getting in there.
My ingenious solution? Schluter DITRA, or a variation for this purpose.
If you've watched any home improvement programs, you're probably familiar with this stuff. If you're not, what it does is create a moveable base between a subfloor and the finished floor that allows for expansion and movement of different underlying materials without the tile or grout cracking.
My idea is to use a system like this between the old and new layers of asphalt when repaving a road, or between the new layers on a new road. As long as the roadway can be made watertight, I would think that the system would allow the top layer to move independent of the bottom layers, to a certain extent. Better yet would be to build drainage into this middle layer so that if water did get in there it could drain harmlessly out instead of puddling and heaving when frozen.
Would it work? I have no idea, I'm not an engineer. But I think something like this could. Sure it would make it more expensive upfront to pave or repave a road, but my theory is that the roadway would last a lot longer, so you make up your upfront costs with lower maintenance costs.