I just PM'd you with my response, but I'm going to post it here should anyone else be interested.
I'm definitely going to give you any answers or research or results, but I'll give you some ideas on how to start:
I would start with a DF target, and get that from a systems level perspective. Contrary to what you think, performance comes last. Integration and the rules come first, IMO. Before you even THINK about how much DF you want and from what component, think of how much the chassis can handle (was it build for Aero?) and how much you'll compromise other subsystems. On our car, the Aero represents a significant load that changes the safety factor a lot. Things like that need to be considered. Find out where you can physically put things before you design them. Again, rules and placement first, performance second.
Start out with where you can mount. If you can mount to unsprung, that makes the analysis on balance very easy to do for the chassis team. If not, well, simplify your models or forget that altogether and start with some simple goals. Our team aimed for 100 lbs @ 40 MPH, which has a big impact on a 319 lb car
You asked for how to pick a CoP and bias? That's a systems engineering task, and something the whole design team needs to design. The same goes for other questions like "how much power should we shoot for?" Sit down with the chassis folk and have a good talk. What do they want to see in terms of dynamics? What kinds of loads can the car handle? How will your target DF's, if met, affect the balance and the handling characteristics of the car?
If you are able, get a very simple lap sim going. Prove aero can work. Use very basic drag models and so forth. Just get an idea (order of magnitude) on what you can expect to accomplish.
Do you plan to run an undertray, front wing, rear wing? If so, target what you want from each, remembering that the tray is infinitely more efficient. Try to use wings as trim devices only.
Typically aero is not popular in FSAE because in order to get noticeable DF, teams think they need 45 degree angles of attack on a NACA 0012. They don't. They need undertrays, and I only know of one school outside my own that does an undertray. They are very, very tough, but if you think you can do it, do it.
Again, integration and the rules are first. Where can you mount? Using endplates, steel supports? Unsprung or sprung? What materials? What manufacturing processes are at your disposal? How many people do you have to build the things? What are the physical limitations on your design space (span, chord, etc.) Don't get thinking on a gnarly wing with twist and changing AoA if you don't have the gantry mill to make the female molds.
As far as wing design goes, start simple. How I started was by just picking a high lift A/F, like the Selig S1223 or something.
Start by doing your basic 2-d analysis. I wrote a panel code that did 5-D optimization on a two element rear wing with a linearly varying vortex distribution, with Thwaites method used to predict stall. In some cases I was within 5% of CFD, even on separated parts. Do some CFD on your airfoil and get the drag polars. If you're newer to CFD, use different viscous models and come up with several curves, and put correction factors based on zero AoA numbers you do by hand using Blasius or something like that. This was probably overkill, but I have a very disciplined and thought out approach on how the cross sections were selected. I personally wanted to justify every design decision and optimize numerically everywhere I could. That's just my approach though.
I do however strongly discourage the "build and test" methods as for an operation as small as an FSAE team, that's just not going to cover enough of your options and is very time consuming and expensive.
That gave me an ideal 2-D section. Then I went to 3-D. First, I characterized the environment by doing a wind tunnel test to survey the wake. That gave me effective Cp's and thus flow rates across the plane I wanted to place the wing in.
I wrote a code that optimized the chord. I set the span at the max the rules allowed, then wanted to look at two things - do I want a really big wing at a low Cl to give me low drag, but high weight, or a small wing that weighed less, but had high drag. Which would reduce acceleration the most?
I set a target DF and then set a multitude of different chord lengths, determine the angle of attack required by that particular size that would meet the DF target, and the resulting drag and weight. I used crude weight estimates at first (some lame 2nd order polynomial I guessed) and then refined them later after wings of multiple sizes were actually built and weighed. I applied the weights and drags to the car and picked the chord length that gave the lowest reduction in acceleration.
Things like that were the thought processes I followed. From there it went in depth into endplate configuration and other things we are still working on. Right now, we're working on aerodynamic twist, using 3-D panel codes in addition to CFD. What kinds of twist functions (linear, polynomial, sinusoidal?) give the best performance and so on. That's the more detailed stuff most teams never touch, but I have the time and manpower to do that stuff.
Hope this helps and isn't too complex, I can tone it down a bit if you have more basic questions.
)
If i figure out where did I found it I´ll post a link 
