Benefits of a properly modeled building?

[matt__w]

I work for a CSMEP firm and we're trying to get some of the less-educated architects or firms that are slow to get on the Revit bandwagon more focused on modeling correctly and get them out of the AutoCAD/Revit hybrid world, admittedly for our own use for energy modeling. We want to be able to tell these architects to model correctly in the first place, but do it in a non-combatant sort of way. I'm not familiar with what architects would use a well constructed model for (besides the obvious... renderings, quantity take offs). Why do YOU need a well constructed model and what could we/should we tell an architect that is only using Revit as more of a 2D tool instead of a 3D tool? I'm looking for advice on how to convince others that they should use it properly.

Thanks for you help!
Matt

[jsteinhauer]

Matt,

This sounds like a tall order, to convince an Arch firm to fully model the entire building. The way we see it, our models are for design intent, where as a model generated by a contractor would be way more precise. We tend to fudge things when developing our documents, using a combination of modeled 3D elements & 2D elements such as drafting components & lines. So find it is not worth the effort to fully build their models out, because it might not be built that way (MEPs more than Architects). If your firm is not the lead on the project (holding the contract with the client), you're going to have an even harder time with this. I would ask that the Architects at minimum enclose the shell of the building, not forgetting soffits & fascia. You're going to have a hard time getting them to imbed the proper R or U values in their elements as well.

Best of luck,
Jeff S.

[matt__w]

Just so I have a better handle on this... what type of objects would you NOT model? You mentioned detail components and lines. For what?

Thanks for the info!

Matt

[damon.sidel]

I think it really depends on the project. Jeff talks about "fudging" things because the model is for design intent. Where I work, we too only do the minimum amount of modeling, but that usually means a heck of a lot of detail in the model in 3d. Depending on the level of development of the project/model, we add enough information for our purposes. If it is too early to worry about wall types, we have relatively generic wall build-ups with a minimum of information about their construction.

What information would you like included in addition to the correct geometry being shown? Are you talking about proper R or U values like Jeff suggests?

We're not going to add information we don't need to add. But if you are talking about organizing the wall and window types, we do that. In our interaction with MEP firms, we've given them our model and they've added the information. To the best of my knowledge, we've given them fairly complete models that they can successfully modify--by adding information--for their use.

[jcoe]

Just so I have a better handle on this... what type of objects would you NOT model? You mentioned detail components and lines. For what?

Thanks for the info!

Matt

I use the approach in our office, if you need to see it in two or more views, then it should be modeled.
This approach has seemed to work well for us and produces pretty decent models.

[Mike L Sealander]

We do some energy modeling ourselves, even though we are an architecture-only firm. It's quite a trick to build a model of any complexity that is optimized for both energy modeling and for architectural construction documents and presentation. That said, the work flow Revit>>Green Building Studio>>eQuest actually works quite nicely. However, I think your main adversary here is the architecture firm that really sees no compelling reason to up their game when it comes to modeling. There are good financial reasons to be that way.

[davidcobi]

I think firms can escape the Autocad/Revit hybrid world when they:
A. Have very well developed templates and content.
B. Have a very good training strategy that transitions any Autocad holdouts successfully.
C. Have management that is willing to invest.

The greatest benefits for us in doing all the work in Revit is:
A. Better coordination between drawings of the same project.
B. Better officewide standardization between projects.

Where modeling correctly is concerned invested training helps, but I think it also helps when complexity is more automated in the templates and content in a way that doesn't require special knowledge on the part of the novice model builder to be able to take advantage of it.

[jsteinhauer]

what type of objects would you NOT model? You mentioned detail components and lines. For what?

Curtain walls, door jambs & heads, trusses, wall base, floor patterns.

I know your first thought, "What 'Curtain Walls'?" Its not that we don't model them, its that we can't have a profile to show multiple closed poly-lines. So what everyone does is they nest in a detail component into the curtain wall profile to show what it really looks like. It would be awesome to have a profile for sun shading, that way it would be as easy as switching mullion types. But that is currently not available. Door Jambs and heads, can be done in the door schedule and details. Having modeled frames might take away from performance. Trusses are designed by the fabricator, so we can't show what they would look like, cause that it out of our hands. Wall base, we just don't do it. Floor patterns, would be great to do but again it falls down to cost vs. benefit. I've seen it done, but mostly in-house it is taken care of in an enlarged plan view with detail lines. We don't want that pattern showing up in every plan, cause it could get very messy for other plan views.

We use detail components in our details, either cut from the model or in a drafting view. We use them for things mentioned above, plus: casework & mill work (there is a difference), structural connections, ceiling/wall transitions, floor/wall transitions, wall framing details...

I hope this helps you Matt. I'm also happy to see that no one jumped all over my original post.

Cheers,
Jeff S.

[damon.sidel]

I'm also happy to see that no one jumped all over my original post.

I don't think anybody jumped all over it because they were probably thinking the same! There is always a point that modeling in more detail does not pay off. Nothing that you list seems all that important to energy analysis, though, so it would be good to hear from Matt some more specific examples. Maybe hearing those examples, we can give him ammunition to go to the architects he works with and show them the benefit to THEM of creating better more complete models. Matt?

[matt__w]

After reading what ya'll have written and having talked to one of our energy modelers, it sounds like having a properly enclosed model is the biggest concern followed closely by having rooms and properly labeled room tags in the aforementioned enclosed areas. A conference room is going to have different requirements than a storage closet. Obviously the more info we can get, the better. Even if it's just the type name telling us what the construction type is. I.E. 8" BLOCK WALL, METAL STUD PARTITION, etc...

[damon.sidel]

I wouldn't even know where to start trying to argue for these very basic things, mattw. The two things you've requested--room labels (tags or even just text?) and a "complete" shell with descriptive wall type names--are so basic that I wouldn't know how to make an argument for them. The only reason I can see that an architect wouldn't model the shell is that it's an interior fit-out or interior renovation.