Yesterday, he wrote a post about the progress of construction on his new restaurant and had a few interesting things to say, which I think reflect the state of our industry. The overall theme of his post is truly about planning and expresses the need for the transformation we are working to make, through VDC. A few great quotes:
Buschel is impressed with the ability of architects to visualize and understand design using 2D plans, because he has no ability to do so - like so many other people in the industry. Only as he stands in the partially completed space, he starts to see "this materialization of a thousand decisions made so long ago in the abstract".
"...So the plumber is at work, playing Twister with the HVAC guy as the electricians dance around the carpenters..."
"...There are drawings upon drawings, and then changes upon changes upon those drawings. Everywhere you look, rubber bands contain the drawings, 24 inches by 35 inches, rolled up like parchment-paper communiqués between Jefferson and Adams..."
I think this is a struggle for so many, who cannot visualize the end space. Once they finally see it,
I would have loved to show Buschel a BIM of his project, sometime during design. As a contractor, we suffer from indecision, late decision, or change of mind. All three are costly and time-consuming, and as Buschel points out, test the paitence of the team. We don't like to see owners struggle to make decisions, and honestly, much of the indecision comes from lack of information. What do the options look like? Cost? What is their impact of schedule?
VDC clearly cannot solve all of these issues, but can help good teams extract and position information to move forward with design and construction.
LEED Platinum Award
The Final Judgement, when an independent judge analyzed our project against our goals to see how well we did (and therefore part of our shared profit!)
Metrics from the Project - the ROI of IPD
The Goal: Slope pipe: 1/4" every 10'
Usually to slope, we just select the entire area that needs to slope and type the required slope. Works like a charm - locating each fitting and pipe at the correct elevation. We did the same thing in this case, but apparently, it is too small of a slope for Revit. Geometry didn't change; slope on pipe reads 0"/12".
No, this slope isn't huge, but the main runs 200', which is a change in elevation of 5". That clearly will affect coordination.
We found a workaround, but it isn't realistic to do this on the entire system, so we've decided just to do it on the main (which has about 5 bends in X/Y). If you change the elevation of one end of the pipe to the manually calculated value, it will slope it (although the elevation still reads 0"/12"). So for each bend on the main, we have to calculate the elevation based on the slope and manually type. (I hope the layout doesn't change too much!) We have decided to ignore the slope on branch lines for now, which run about 50' - only because we don't have to have to do so much work if things change.
This really shouldn't be a problem. Are we missing something or can Revit really not recognize slopes below a certain amount? Is there a better workaround?
On this project they were only design team member to work utilize VDC (although now they are working with Revit-enabled architects on about 20% of their projects). The contractor, a JV between Austin and Walsh, utilizes BIM, but initially wasn't sure if they could trust the accuracy of LA Fuess' model.
In the end, LA Fuess proved that their model was accurate and detailed enough for use in coordination and fabrication-assistance. They actually added geometry (like gusset plates) at the request of the contractor (a lesson to all of us: ask and you shall receive).
The model was transfered to the fabricator for detailing, and the information was so beneficial to the fabricator that it cut down the detailing schedule by 2 months.
I was really impressed with the level of modeling that SGH did - every stud, framing, etc. I know from first hand experience that wood frame modeling is somewhat painful. Some of the issues that were brought up related to the documentation of the information after it had been modeled (tagging, line weights, etc.), but we definiately ran into some of the modeling issues - there aren't a lot of families and everything has to be modeled manually.
Hopefully life will improve with either the 2010 enhancement tool or the latest version of MWF (which I've heard is quite nice!).
It looks like Revit exports direct to ASD and then users can use detailing macros to add connections, piece marks and create shops. I don't believe there is a bi-directional workflow, but it seemed like updates can be sent from Revit to ASD.
I noticed a "formwork" button, but Derk told me that functionality is not quite ready. Hopefully soon, as we have a massive CIP project we are starting! We have been looking at Tekla for this and more, so it will be interesting to compare the two platforms. Although functionality is key, we're also interested in going with the product that most fabricators use. Collaboration is a huge driver of our software decisions.
Here are my notes/thoughts on the webinar (copied from the Q5 blog..do you really expect me to write a unique post?)
Today, the BIMForum Constructor’s Subforum held its first monthly webinar, discussing laser scanning. The BIMForum decided to tackle this topic because, in the words of Pat Carmichael from HKS, it enables a “level of detail of built analysis that you can’t effectively get any other way”. Pat gave the architectural perspective on laser scanning. He walked through several of HKS’ use cases, mostly stadium projects, and addressed the value that they get from the process.
Mario Mazzi, from one of Turner Construction’s West Coast offices, outlined their process for planning laser scanning. It is similar to one that we would recommend, so we added some Q5 advice in context:
Step 1: Plan
Determine the end use for the laser scan data. Do you want to coordinate new design with existing conditions, compare as-built conditions with as-designed data or something else entirely? Will you convert to BIM or use in native point cloud format?
Step 2: How to Collect?
You basically have three options: subcontract the laser scanning, rent a laser scanner or purchase your own. Some general criteria:
If you have over 6 months of laser scanning to do, buy. Why? Because laser scanners cost $120K to purchase and $20K/month to rent.
So how do you decide if you want to subcontract or rent? It really depends on your organization and your project. Do you have surveying capabilities in-house? Can you dedicate a team to laser scanning? Do you want to train staff in data collection or modeling? How large is the space? What type of scanner will you need? Since many projects require multiple types (for close-up and distance), it may not be cost effective to rent.
Step 3: What to Collect/Model?
Determine your level of detail. What objects do you really need to capture or model? Although a laser scanner will capture everything in “line of sight”, you may not need to convert everything to Building Information Model format.
Step 4: Map Your Schedule
Create a schedule that includes time to capture data, which may be based on when you can access the space. Map out how and when data will be sent, and if data will be sent intermittently or only when complete.
Until now....? I just got a marketing email for Navisworks Surveyor, a plug-in that organizes exports from Navisworks based on Search Sets. I could see this being very helpful, as we use Search Sets ad nauseum to organize our models. Once we set up a file, we could extract quantities directly from the Search Set, making it easier for the cost engineer to understand "what the quantities are coming from". (Although, yes, they could do this with QTO, too.)
This could definiately expand our supers' capabilities. If they need quantities for ordering, they can just extract them!
This looks pretty interesting, even though the interface is a little tough looking. And until December 31, they have a BOGO sale for Navisworks customers. I'm still digging into it, though. Is anyone using this?
JC Cannistraro focuses on the ROI of VDC in their Fall 2009 newsletter. The summary: the more they utilize BIMs, the more they reduce change orders. Furthermore, social BIM reduces change orders even more.
Makes sense, but the ROI needs to be pushed to the next level. What about the ability to reduce schedule and budget (baseline budget, not change orders)?
We use DesignReform's tutorials a lot for training, so I'm happy to add this one in! Thanks David - now you just need to explain View Range for RCP.