We all know the answer is “Yes!”
In a perfect world, quantity extraction would work with any model and team. However, this is not a perfect world and automated estimating does not exist!
I must admit that at one point in my career, I thought we would get there. I still hold out hope that we will—Just as I hold out hope that we will see flying cars. Or better yet, we will mass-produce Picasso paintings that yield the same money as the originals.
Can you automate art?
If we argue that cost estimating is part art and part science, then doesn’t it follow that only a portion can be automated? The algorithm for creating freehand expression has still not matured. Sure, we’ve automated many of the tools that help us create art, but it still requires a human being (artist). The same is true of cost estimating—we still need estimators (human beings) and will continue to need them for a long time.
This is how I see it:
- Feasibility (PreBIM) stage. In most cases, the model has not even been created at this point. It’s usually just a discussion in narrative form with folks that have never built anything before and maybe there are some spreadsheets. Tell me how a quantity extraction tool helps me here?
- Schematic Design (SD). Often no model exists or it contains very limited information (data). Seldom do we have a constructible quality model in any form. Quantity extraction usually provides limited value.
- Design Development (DD). Now we’re talking. Give me your quantity extraction tool. Wait… what? You’re telling me we don’t have the information we need in the model? So extracting useless information doesn’t deliver any value? “But look, I can import it into the estimating solution,” said the VDC expert that was escorted out the door and asked not to come back. And by the way, the owner wanted a guaranteed maximum price at the schematic stage.
- Construction Documents. This is where the model meets utopia—or is it? Are you kidding me? You’re telling me the model is still not there? Then what do I to do with this quantity extraction tool? Oh, I can take off the structure and some of the architectural elements? But I can only use it to check my work? At least that’s something, and it does justify my one seat of [insert product name here].
- This is also when many say that the integration of 2D digital takeoff and 3D should happen. However, that is still up for debate. Many people get lost trying to figure out where the information resides which has led to massive double counts.
- Once you have the perfect model, it’s typically six months past handover. For all the facility managers reading this post, please bear with me. We are focusing on models in preconstruction at this point. We all know it is too late. Even though we have the ability to extract all the quantities (data) from the model, our time has come and gone.
This is why I believe we need to redefine cost estimating and BIM (5D). Think about the advantage of having a project history system tied to scope that is defined coming out of the model. This would let the rest of the process and tools mature, which is the right thing to do.
Yes, there are exceptions where schematic models are reasonably rich in geometry and even data. The key word here is exceptions.
Sorry I went long on this post. Thanks for hanging in there with me. I’m interested in your thoughts.