There was a discussion of this in a recent issue of ASDSO's monthly publication. I'm out of the office so, don't have the publication name or issue number handy.

In short, the factors of safety can be different since they represent different phenomenon. However, criteria are also usually quite different.

Personally, I use the method described in EM 1110-1-1913 by USACE for the design of levees. In that method the gradiant accross the impervious statum is compared to 0.5 for materials with a moist unit weight above 110 pcf.

Mike Lambert

Equation FS=ic/i will give safety factor with respect to critical exit gradient against boiling or heave. This failure mode is only applicable to the conditions where a levee is founded directly on cohesionless soil.

The safety factor against uplift is more applicable for your case. The confining layer(impervious stratum) may experience blowout if the seepage pressure in the pervious layer is higher than the overburden pressure of the confining layer .

I don't believe ic/i should be considered a factor of safety against piping. It's really about vertical seepage forces overcoming gravity, and Rilxouss is correct about it being for boiling or heave. Piping can occur from upward, downward, or sideways seepage. Some attempts have been made at finding the minimum gradients that can cause piping piping - John Schmertann, maybe 15-20 years ago, and somebody at Technical University of Delft working with clean fine sands. See if Google will find you anything. Very dependent on material types.