This may seem like an off-the-wall question but I'm going to ask in hopes of getting a few opinions.

Long story short I have a project that involves modifying an existing embankment slope that was constructed over very soft clay (Su values in the 100-150 psf range). The embankment modification includes flattening the sideslopes to go from an existing 1.5:1 slope to a 2:1 slope. The issue we're running into is that the underlying clay material is too soft to support the wedge of soil we plan to add to the embankment. Our slope stability models predict a very shallow slope failure will occur. The slip plane roughly parallels along the top of the existing slope, penetrates a short distance down into the clay layer, and then exits out of the clay layer 5' to 10' beyond the toe of slope.

We've considered using geogrid in an effort to improve the factor of safety against sloughing/sliding. In the past I've always seen geogrid placed in a series of horizontal layers. However, doing so would require excavating a good portion of the embankment. After thinking about this a bit I'm wondering if placing a few layers of geogrid parallel to the top of the existing side slope would work just as well. The geogrid could be anchored at the top of the embankment beneath the roadway fill and then laid out along the surface of the existing side slope. The proposed fill could then be added over these geogrid layers. I'm thinking we'd also want to loop the geogrid a short distance back into each fill layer to create a reinforced "bulb" of soil at the base of the embankment. In this way the geogrid layers would act as tension elements to resist slippage of the slope and the soil "bulbs" cold help support the slope above. I recognize this may be an unconventional arrangement but it seems logical to me from a load path perspective.

Does anyone have any thoughts on such an installation? Even better, does anyone have any references or case studies for such an installation?