by Mark Sawyer
Why not start with Controversy #1? From the owner’s perspective, if the architect is BIM-savvy and produces a 3D model as part of his workproduct, why can’t the general contractor and subs use that model to plan construction and fabrication? Or more to the point… why should the owner pay the GC or the subs to make another model(s)?
The short answer is one model can suffice… but it seldom does today. Why? Too many parties with too little shared expertise and near-zero economic crossover. In other words, disparate functions, disparate knowledge sets, and disparate P&L’s.
But, there is precedent for making something akin to “model sharing” work. The subs have for untold years “re-modeled” the design. The entire practice of shop drawings is a re-modeling effort of sorts; never mind that it’s usually 2D drawings and not 3D models. It’s an accepted loop in the process that exists so that a mechanical sub, for example, has the data he needs to fabricate systems and components. No one would dream of telling that mechanical sub that he can’t author his own data – maybe a model – in order to manufacture the requisite elements. The owner doesn’t insist, “No, no. You can’t make your own drawings or build your own 3D model to fabricate that ductwork. You can only use the drawings handed to you by a design firm you haven’t met yet.”
There exists a precedent whereby all parties just know that the HVAC guy is going to generate shop drawings – e.g. some new data. I haven’t heard of an owner balking at the cost of this effort. And, after all, those same shop drawings are reviewed by the design/engineering firm(s) for accuracy and completeness, so there is no foul.
So I’m going to narrow the question. Why, then, is it controversial, or redundant, or unnecessary cost when a GC sets out to build a model that reflects the means and methods by which the building will be constructed? If that GC wants to extract quantities (and I don’t mean window schedules, or door schedules. I mean the kind of quantities from which he can derive materials for formwork, or right-size a rebar crew) he is going to need a model that delivers the right means and methods data. If that GC is going to compute the duration for pouring each of four sections of a slab, he’s going to need a model that breaks one big slab into four individual segments. The examples abound.
Admittedly, the architect’s model could reflect the means and methods data needed by the GC, but that is expecting too much of the design firm taken in the norm. (I’ve seen exceptional examples to the contrary on projects in the US and abroad, but they remain the exception. And they have each involved alternate delivery contracts or an international variant of same. Hold that thought, please.) It’s certainly not included in the architect’s fees. Nor is it expected that the architect would assume the cost risk associated with those derived numbers. As a result, means-and-methods is not a cornerstone of the architect’s knowledge nor her perceived value on the project team.
The GC who derives quantities, costs, sequences, and schedules from a 3D model (the 5D Virtual Construction approach) requires a means-and-methods model. Whether he authors that model from scratch using a designer’s drawings, or whether he takes the designer’s 3D design model and embellishes it… he has to arrive at the same endpoint. In so doing the GC is not revealing a failure on the part of the architect or engineering firms. He is simply fulfilling his own data requirements (just like those shop drawings for the fabricator) and thus perpetuating the BIM process through construction.
Those are the facts around data and models. Any decent modeler with foresight into the intended use of the data (e.g.. coordination, quantity take-off, model-based cost estimation, sequence animation, or model-based scheduling), and working knowledge of those uses, can produce a good BIM. We would all do well to make sure the owners understand and are prepared to embrace these facts. (And, yes, that includes embracing the notion of paying for multiple models when necessary!) Where this gets tricky, and I’ll argue where this all breaks down into unhealthy contention, is in the contracts and their interpretation: a.k.a. politics. That’s when we all change our tune from “PolyModel Doodle All the Day” to “Nobody Knows the Trouble I’ve Seen.”