What is the best way to model a drop in the steel framing for a slab depression?

Sorry. The middle bay should be -0'-5".

There is no convenient way to do this exactly. Assuming that the girders are still braced by the infill beams I say don't do anything about it, just leave the framing all at the same elevation, and make the change on the drawings. If you are concerned about connection design, you can run our built in connection check or use Ram Connection and adjust the elevations there. If you want the deck to be thicker in this area, just assign a new, thicker, deck to that bay in Modeler. 

Thanks Seth. I have a few followup questions:

1. The condition in highlighted in yellow has a WT welded along the partial length of the beam to support the higher portion of the slab. Do we model these as completely noncomposite? 

2. How does RAM recognize the effective slab width at the locations highlighted below. The green condition has the depressed area on one side of the beam or not spanning the entire span. Does it still check it against a) 1/8 beam span b) half distance to CL of adj beam c) distance of edge of slab as specified in AISC?