How come she doesn't come forward and ask the question then? They are several directions 'she' can explore but the description is too vague. Is this an undergraduate thesis?

Sounds like you (sorry, your friend) are well on your way to accumulating the requisite amount of buzzwords for a thesis.

Just throw pedagogy in the title somewhere and you'll be golden. Maybe something like:

"The Pedagogy of BIM in a Lean Integrated Design Context for Reinterpreting the Contemporary Urban Fabric of Favelas"

Or PBIMLIDCRCUFF, for short.

The Pedagogy of BIM in a Lean Integrated Design Context for Reinterpreting the Contemporary Urban Fabric of Favelas"

 Well I did something like that only is was the Octavia/Market/hayes Valley area of San Francisco and found the urban void there -

I did my thesis on BIM - I devleoped a design process using that goes from Urban Analysisof the immediate zone)a 5 block region(Hayes Valley), where I used used parametric families to represent the typology, then use the data from that to drive the Concept Design>Schematic>DD>CD

I did the entire process in 1 architectural file, and 1 linked site file - Any concept change made anywhere in the pipeline, can be quickly made downstream - as I produced 10 variants in a process that lead up to a reduction that produced the final resultant design.

where I used used parametric families to represent the typology

Did you actually use ArcGIS, you know, to actually create an urban analysis? You know where you link a city's building shape file to the TIGER data so you can geocode FactFinder? That way you can actually represent the income levels, number of units, people per unit and use the wayfinder to calculate where the best locations would be?

I'm pleasantly entertained to find out that my mad libbed archibabble title had academic "merit" behind it.

I think it might be more interesting and more profitable to write about what BIM isn't rather than what it is.  

There's reams of developer propaganda that extols the virtues of a fully integrated BIM project, yet most projects are still done with hybrid workflows.   Often, even if both the architects and lead consultant are using BIM, they're using different platforms which are incompatible in terms of integration.  

Nevermind that Bentley Architecture still needs a lot of work, Revit still can't really handle large scale buildings, and bespoke complex geometry is difficult/annoying/tedious to implement, Digital Project and Catia have somewhat steep learning curves, Rhino BIM is still in the beta...

I would be interested in reading a thesis that explored the following issues:

1  Integration of building information across platforms.  Can a Revit model ever be truly dynamically linked to a Catia model? If not, why not?

2  Deliverables - An exploration of the legal ramifications of a digital model deliverable in place of printed drawings.  Can the architect and GC work on and discuss the same exact model, legally?

3  Software and hardware limits - Where does one 'draw the line'?  Meaning, exactly where in the workflow is it faster and more logical to draw in two dimensions rather than modelling?  If details are drawn in 2d and not modeled, how does that effect quantity surveying and cost estimating?

I could go on with more pitfalls/questions/issues, but I think those are the most important to solve ASAP.  

So my suggestion for the title would be something like ''Building [mis]Information Modelling''.

And, as you posted on archinect asking for advice, I would hope you'll follow up and link to the final product.