When Systems Collide
If you’re like me (and let’s face it, if you’re reading this article the day it came out then you ARE like me), you’ve been looking ways to best approach the problem of M-E-P clash coordination. I’ve seen multiple walkthroughs and guidelines, but most of them don’t really apply to what I am typically tasked to do. Frankly, I don’t have the time to do it the way they say to do it.
In a Perfect World
The articles I’ve read begin innocently enough. First you combine files, then you create Clash Sets, then you flag the clashes by responsible party, then your subcontractors address them. See you next week.
But when I add the difficulty of low floor to floor heights, increasingly complex systems, high ceilings, and a short schedule, I end up with THOUSANDS of clashes per week!! There has to be a better way. This article attempts to address these issues with approaches to make best use of your short amount of time—without flagging 15,000 clashes per week.
The fact is, with six full-time detailers providing me with new models on a (at least) weekly basis, I can expect this amount of clashes at the beginning of a project for, at the very least, several weeks. For me to do my job the way I just described would take days, and I don’t have that kind of time. So what to do?
Figure 1: An overhead example of a simple coordination.
If Figure 1 represents what one of your project typically looks like, this article may not be of the most benefit to you. For everyone else, read on!
Reality Sets in
When you first start a project you might take the model you got from your architect, or perhaps a model that you created yourself, Export out an .nwc file (the Navisworks Cache File) from Revit, and Append that .nwc into Navisworks. From there, you can Append relevant .dwg files that contain additional information, and then it’s time to share.
You will use a file sharing service such as a dedicated FTP account that your IT department may have set up, or perhaps you use BOX.NET, Dropbox, or a similar service. You can share your base files with your team, and decide on a 0, 0 location so that everyone is coordinating in the same space.
Next, you need to Append their files into your base .nwf (this is the Navisworks file format for combination files) and verify that all of your files are lining up.
Next, you need to set up Clash Sets. But what is the right approach? Do you work in a hierarchal setup, where your first clash set is Fire vs. Everyone, then Drainage vs. Everyone except Fire, then HVAC vs. Everyone except Fire and Drainage, and so on? If you work this way, you may find that some of your Clash Sets come back with thousands of clashes that have to be sorted through.
It is for this reason that I work in a true “trade vs. trade” clash environment, regardless of responsible subcontractor. This gives me upwards of 30 clash sets per project area.
This is the time in a project when I am tempted to tear out my hair. With all of these clashes, there is really no way to address each one, but in reality, there is no reason to do so anyway.
Your job at this point isn’t to point out each individual clash, your job is to identify pinch points, or areas that don’t look like the designed items are going to fit. Leverage your subcontractors’ experience and knowledge of their own systems to find out what they can and can’t do reasonably. Discuss major ductwork crossings, and find out what can be done to resolve these issues. Find the general direction the team wants to go, by working together. Remember, nobody wants to model their entire system, only to find that somebody else put all of their items at that same location as well.
Figure 2: An overhead example of a complex coordination.
Once you have determined the basic direction for the team, you can start your coordination by stepping back and flagging the specific areas of conflict. Then you can try and decide on the approach that you are going to utilize to resolve these issues.
Often you will run into an area that has lots of repetitive clashes; arrayed conduits or pipes that run through ductwork, where hundreds of clashes are found. In reality there is only one issue to address. To address this, I suggest leveraging the view settings in the Clash Detective. First, use a selection window to select the offending objects. (To minimize the items you have to select I would pick the ductwork in the example above.) Then use the Filter by Selection option in the Display Settings, then New Group the clashes, and assign the party. Name it something like HVEL 01. Using this technique throughout a project can reduce a Clash Set from having several thousand to a much more manageable several dozen or so. Use the Report functionality to publish Group Headers Only to HTML to reduce the apparent amount of clashing in the project. By exporting the report using the Group Headers Only option you will only see the settings contained in the top level Group of each clash.
Figure 3: Combine clashes into Groups.
By using a Template File with which you start each project, you can avoid the tedious task of setting up Clash Tests at the beginning of each project. These tests should include individual clashes, or “Trade vs. Trade.” By analyzing the project in this granular approach, you minimize the volume of clashes to be found by any one Clash Set.
Not only should you build your Clash Tests to clash Trade vs. Trade, but you should also include several extra tests of All Trades against things such as Structural Steel, King Studs, Top Track Clearances, and Ceiling Grid. This will verify your subcontractors are installing as per the construction documents, along with the added benefit of notifying you of any discrepancies between the Consultant Drawings versus the Architectural Drawings. (A-100s versus M-100s versus E-100s, etc.)
If the Clash Tests in your Template file are built by utilizing the Sets Tab of the Selection Tree, you can automatically have all of your tests populated just by verifying that your subs use the specific rules that you will build to create your Search Sets.
In my process I have found that by enforcing a strict naming convention on our subcontractors, we are able to streamline many of the processes we utilize in Navisworks. For example: we make our electrical subcontractor name their file something that contains “EL” in it, and it doesn’t even matter the file format. You can then use the “Find Items” tool, look for the specific “EL” term in the name of the file, Find All, then save this as a Search Set. Now you can always have a Search
Set for each trade, even as they update their files each week. (You can even export these Search Sets to apply to other projects.)
Create an individual Search Set for specific items such as Diffusers, Recessed Lights, Ceiling Grids, King Studs, Structural Steel, and any objects that could set you up for smoother results down the road as well.
Figure 4: How to structure clash sets using search sets in a template file.
For example: when clashing your “HV” Search Set vs. your “EL” Search Set, you can create a Rule to ignore items in Search Sets “Diffusers” or “Recessed Lights” using the Custom Rules function. This will pull out items that are supposed to be 1” or less apart from each other, (the registers and the lights in the ceiling grid), and allow you to concentrate on the items that are actually an issue.
To address the possibility of lights and diffusers actually being in conflict, create a Clash Set between only the Search Sets “Diffuser” and “Lights” using a Hard Clash with a 2” Tolerance. Using this same technique you can clash the locating of these items in the ceiling grid with a “Ceiling Grid” Search Set versus both the “Lights” and “Diffusers” Search Sets.
Navisworks Manage has brought some new functionality in its 2013 release. Among these are the ability to natively import Revit files, the ability to display grids and levels from Revit, and a totally redesigned Clash Detective tool.
Although I have not implemented the native import of Revit files into Navisworks into a real project yet, I have sampled it, and it looks like it may be the way I approach coordination in the future. One benefit is the way that the Selection Tree separates items by object type as opposed to by hosted level. This is perhaps a personal preference, but I find it easier to navigate the model through the object type format. One drawback, however is that it does seem to take quite a while to import the Revit file, so the complexity of the project may keep this from being an option for you.
The Grids and Levels display functionality is a great way to orient yourself throughout the project, and is also available as a column header in the Clash Detective, making it easier to locate clashes throughout the project.
A previous article covered the basics when it comes to the new Clash Detective, so I won’t go over it in detail, but I will say that being able to hide both the display settings and the items clashed, gives you more room to understand the items that are clashing, making clash detection that much easier.
If you haven’t taken the time to embrace the Appearance Profiler tool, this is something that I would recommend as an easy way to simplify communication between trades. When you look at a file that comes in natively from your subs, it may be hard to tell if that round pipe is an electrical conduit, or a domestic water line, or a med gas line. The colors that have been predefined by your subs just aren’t that easy to lock into your brain. (I’m sure it makes sense to them.)
So take some time to build an Appearance Profiler file (as a .dat). Use the Search Sets by Trade that you have built previously to assign a new appearance to each system. Use what makes sense to you. You could choose brown for waste/ vent, blue for domestic water, red for fire protection, etc. I even go one step further and separate out my mechanical systems to supply, return, and exhaust. (Most third-party detailing software provides a field in the properties of the object to determine to which system they belong.)
I’m sure you can imagine the communication that can be streamlined by using this approach. Now every time I open a file, I can confidently say whose items are where.
Figure 5: Appearance Profiler—before (left) and after (right) comparison.
Using the Appearance Profiler has the added benefit of showing you exactly what items have been added since your last meeting. As new items are added to each appended file, they will come in as their original color; you will then be able to identify these items giving you a heads up on what has changed. (Of course you will then want to rerun your profiler.)
One of the most worrisome things that I’ve ran into while using Navisworks Manage 2013 is its tendency to crash. I’ve ran into multiple tasks where, if you use them, you’re in for an error report.
Figure 6: Not a graphic you want to see often.
The first, and most problematic for me, is the Refresh command. Simply hitting the Refresh command crashes Navisworks (for me) about 60 percent of the time. I now prepare for this, and save immediately before doing such a refresh. (Sometimes I just close the file and reopen it.)
The next major issue (bug) comes when Grouping clashes into clash folders in the Clash Detective tool. (Or dragging additional clashes into existing clash Groups.) Utilizing this workflow can cause Navisworks to crash on a consistent basis as well. Using the process that I described before, you can see how this would be especially problematic as that is how I approach large Clash Sets. The only work around that I have found for this one is to be deliberate when combining clashes. I’ve found that if I’m trying to work too quickly, that is when the crash will happen. So I slow down, wait for my 3D view to settle down, and then drag the selected clashes into a group folder, (or create a new one).
Both of these issues have me doing something I haven’t done since AutoCAD 2006; saving often, ramping down my Autosave timings, and being ready for work to be lost. Hopefully this is something that gets addressed sooner than later by Autodesk.
Embrace the Designers
As your coordination process continues, one potential bottleneck that you may run into is the hesitance of your Subs to make revisions without the appropriate RFI process being employed. This keeps them protected from rework, and progressing too far down a proposed solution, that may ultimately be the wrong one.
One way you can help them move quicker is to reach out to the design team early and get them to buy into your process. I will typically get introduced to them through my project manager, and ask if they are willing to meet with me on a weekly basis, for about an hour at a time. They are usually excited to see such planning to streamline the process.
During your weekly coordination meetings with your subcontractors, take notes of items you want to bring up to the design teams; the architect, and consulting engineers. These items could be the sizing or routing of ductwork, the option of lowering the ceiling in specific areas, missing dimensions, missing piping sizing, and others.
If you can bring these items to the table quickly, I prefer the meeting with the designer team to be the day after the M-E-P coordination meeting; you can get answers immediately about how to proceed. Of course you will need to have your subs follow through with the appropriate RFI process to tie up the paperwork side of things.
At the end of the day, your job is most impacted by your ability to communicate effectively. It’s your job to resolve issues early, without feelings being hurt, and in a mutually beneficial manner for all parties involved. If you can’t do this, you’re in for a long coordination process. Remember that everyone is on the same team. We all want to finish the project on time, under budget.
Josh Taylor is a Project Manager of Virtual Construction for one of the largest General Contractors in the Mountain West. Josh has seven years of BIM-related experience in both the Architectural and Construction fields, has worked with AutoCAD since 2002, Revit since 2007, and Navisworks Manage since 2010. Josh can be reached for question or comment at firstname.lastname@example.org.