yeah i like this idea.  the way i would handle it is this.

if a pioneer builds on say a fertile land tile.  without ANY caravans at all you still get what is in the game now, i think that is half of the resource production.

however, if you had a caravan going back and forth to the tile(like it was a mini settlement) then you would get the half it produces plus whatever the caravan could carry as well.  maybe the caravans would be upgradable, or better yet DESIGNABLE so as to carry more, or perhaps have multiple caravans transporting goods.

Another option would be to enhance them in a way similar to starbases from galciv... multiple pioneers establish walls and defences, caravans boost production up to city levels and you can station troops there.

I think the problem is that it would be much faster to use pioneers than city building for any uncontested regions so you can't just boost production to pioneers or rushing pioneers would become overwhelming. Caravans at least take time, having them upgrade over 'n' turns based on outpost level could work... 1 pioneer will upgrade every 50 turns from 1/4 production to 1/3 to 1/2 to 1/1 an outpost level 3 would upgrade every 20 turns and reach a max of 2/1... but each level requiring greater levels of research.

Pioneer only resources would be interesting... something too dangerous to build a city adjacent to (volcano tiles for the kingdom, 'ents/fairy grove' for the fallen) You want to promote battles over certain resources by making them more vulnerable and reduce wars started over a tiny resource. (Especially if the pioneer and city bonus for each race made holding that spot desirable... the kingdom wants to build an outpost on the volcano for a magic bonus and res extraction and the fallen want to build a city for troop bonuses and magic gains and a unique building)

yeah, i like it.  the only downside i see is slowing the game down from a hardware prospective.  if you have 30 or so caravans moving around and so does everybody else, thats alot of units to move each turn for the CPU.

The problem with this approach (as opposed to having to generate multiple caravans at a city and sending them to the outpost) is that there is no decision behind it.  You will ALWAYS use the caravan and just get the 100% outpost production.  I think the devs want outposts at less than 100% to simulate the remoteness and the difficulty involved in moving large quantities of materials over distance in a medieval society.  

Having to generate the caravans yourself and send them to the outpost in order to increase its productivity beyond 50% means that the player is will to commit time and resources to getting the extra production and maintaining it (through protecting the new trade route).  It's a strategic decision at that point rather than an automatic one.  

I understand your argument and why people like the idea of tying tech level to outposts, I just don't like the idea that there is no investment or opportunity cost.  In you way, as your empire grows and tech level increases you will max out every resource outpost automatically just because it is trivial to do so. There is no barrier to entry or investment required.  All that it requires is maintenance cost (protecting the route) which, in theory, as your empire grows you'll be doing anyway.  

I think that upgrading outposts from 50% to 100% works best if it isn't an automatic choice.  Say you have 4 ore mines attached to cities, but have 3 others within the "heart" of your empire (the part most well guarded).  Throwing a pioneer on them and just letting caravans and tech research up them to 100% is a no brainer, because the protection for the routes is already there so the extra production doesn't cost you anything.  You're decent tech level gets you the extra 50%.  So essentially you're getting (in this example) the extra 150% metal production for nothing.  Your society has reached a point where there is zero cost (aside from the trivial pioneer) associated with getting all the production possible.  

If you have to create and send multiple caravans, then there is an entry cost to go along with the maintenance cost (protection).  Your need of that extra 150% production (in the example above) matters because it isn't free.  If you're raking in more metal than you can use from the 4 ore mines directly attached to your city and the three pioneers you just sent out, you probably wouldn't upgrade your mines because it takes money, resources, and most importantly, build queue time.  Whether or not to upgrade outposts is now a strategic choice, because you have to spend real resources specifically on it (as opposed to it just being something that happens in passing via tech level).  And now upgraded outposts (with their fancy, expensive supply routes) are even more important to protect and/or attack because they are very much a non-trivial cost.  Their loss, not just the production but the investment they represent, is a big deal now.  In the example above, their loss isn't huge.  Just secure the area again and send out another pioneer and you're back in business.    

You'll probably still reach a point very late game where upgrading those outposts is largely trivial because caravan production is so fast and cheap (Relative to your empire's economy), but that point will arrive much later in the game, making upgradeable outposts a actual decision through a much greater period of gameplay.  And largely trivial isn't the same thing and a fire and forget pioneer (because tech level handles the rest).