I know this is an old thread, but thought I would add some thoughts. There are several issues. As a caveat, I speak with some knowledge of Okinawan kata, but cannot extrapolate to Kempo or Chinese arts.
First, your effectiveness against multiple attackers will be significantly related to how quickly you can dispose of the initial attackers. The longer it takes to stop the first one, the greater the risk that attacker 2 or 3 can launch a successful attack, especially from your blind side. Also as your "engagement" with attacker A drags on, the risk also goes up that two simultaneous attacks will occur, which puts you at really great risk, especially if one is an attack to your legs with a one or two legged takedown.
So if your can't stop the first attacker in a reasonably short period of time (or at all), then the whole issue of multiple attackers kind of becomes moot.
My biggest concern with the vast amount of bunkai I have seen in a variety of karate systems is that it would not be effective against a large attacker. I will get on my soapbox here. A surprising amount of bunkai isn't just "not good". It's bad. Bad fighting principles are used.
I have trained in systems that practice great ideas in non-kata related self-defense. They get off the line, use multiple counters, often to the neck, there is good body mechanics to maximize power, and many combinations include takedowns. Just plain good fighting. But then they do their bunkai. They don't get off the line. Blocks are used in ways that they would likely not work well. And so many of these defenses launch a single counterstrike to the solar plexus.
I call this single strike defense the "arrogance of karate." I have had the good fortune to have had exposure to lots of different systems (Chinese, FMA, MT, JKD, Kempo) and this principle isn't found elsewhere. Why? IMO, against a large attacker, this is a good way to get really hurt. Arnisador, in the opening of this thread, mentions the one shot-one kill approach, but against a larger attacker, there is just an unbelievable amount of risk with that approach. Especially because the target of choice is most often the solar plexus, something that is really hard to hit on a large fast-moving attacker. If you miss, which is likely, you hit the upper ribcage, sternum, pectorals or abdominals. These are not exactly vital targets. In reality, these are just about the worst targets to choose on a large attacker.
So you have to start with meaningful interpretations. Let's assume we have them. We have multi-counter combinations that are designed to really stop a big guy.
Regarding Okinawan kata, there are lots of techniques that help against multiple attackers, but I will discuss two general principles that are based on the footwork of common turns in the kata of Shuri te and Naha te. When turning in Naha te (Goju) kata, it is common for the front right foot to step across the back left foot to turn 180 degrees. This movement gets you off the line of an attack from the rear. If you can't see the attack, but you have an idea it is coming, you need to take a guess at what it might be and how you might respond. I would argue that a right strike (probably one with some arc.) has the highest probability. So by stepping with your right foot, off to your left, you are stepping off the line and away from a right strike you assume, but can't see.
Also, by pivoting counterclockwise, you set up combinations where you block with your left and strike with your right, all on the counterclockwise turn. You bring good body mechanics to both the block and the strike. (Assuming they are near simultaneous, as they probably should be.) And if you are right-handed, you might like this setup of your right hand to be used for the counterstrike.
Let's compare these Naha turns to those found in Shuri te (Shorin Ryu) kata. Here the turns are commonly done with the back foot, and some of these turns go up to 270 degrees with a counterclockwise spin. This wonderfully complements the Naha approach. The turns become throws and takedowns, which are commonly done with big spins requiring the backfoot to step. Now you can use these takedowns to insert wounded attacker A in the path of attacker B.
Think of a fight as a game of chess or pool. For those of you with skill in either of these, you know that you don't make moves or shots in isolation. Moves and shots are taken to establish good position. They set up further moves and shots. The same can be said of these movements in kata. One might choose technique A because it puts an attacker from one's right, over to one's left. If there are two attackers, that would be a good option for an attack coming from the right. Or if there is someone behind you where he can't be seen, then you might choose a technique where the takedown puts the attacker 180 degrees to the rear.
But here we are getting into the realm of true mastery of kata. We need to be able to add a new dimension to our choice of defensive combination. You still have the usual considerations. (what is the attack?, how big is he?, how skilled?, what's open?, etc.) But now you add to that what the next risk is. (e.g. of the three attackers, after the first, which guy is bigger, more aggressive, or can't be seen.)
There are plenty of other ideas on this concept of multiple attacker defense that I can share if there is interest.
