After testing this application last summer and more recently with Beta release version 1.5, we offer the following brief review. To view our instructional video, click on this link:Assemble 1
This is one in a series of studies, of a dozen applications used to more accurately extract and filter data for estimating. The secret to this tool is it’s simplicity; it’s easy to use. This study focuses on estimating from models that were furnished by outside design entities, rather than by control models built specifically for estimating.
The challenges with any 5D extraction process is having mechanisms in place for verification to be able to trust what is being derived.
The problem with modeling is the time it takes to setup schedules within the design model, to be able to readily extract from when the design is influenced by many that are not under the control of the estimator. Various quality checks are needed requiring the experience of a Senior Cost Planner when receiving “Live” design data that is in a constant state of flux. A recent example of this was discovered when extracting concrete volume from a structural engineer’s model for a mixed use, hi-rise project. The model transmitted to our office was an analytical model lacking the detailed geometry called out on the architectural sheets. The concrete slabs around the perimeter bays were indicated as 2” thicker resulting in the concrete quantity to be greater than the conventional estimate, thus causing concern. The problem was easily resolved by having the structural team send us a detailed model the following day, but this points to the need for model verification. Having a reliable system of checks and balances is essential.
Today models often have hidden information, data that often is not on a the printed set of bid documents. But this hidden data is difficult to get to. To pull out this information not only requires training and experience but also a deeper understanding of how the information is organized. This is the niche where Assemble has focused their application.
Assemble functions as an export tool from Autocad and revit, pushing data to their extranet, where users login to sort the data and filter what they want, before creating excel reports and selection sets for viewing what they just sorted. After this sorting process is put in a format to match an estimate, the data is then downloaded adn the XML set loaded into the viewing application, Navisworks. A user with access to Navisworks Manage can then export a file that matches the Excel extraction, viewable with Autodesk’s Freedom software. (A free viewing tool)
Assemble was tested on mid-rise projects, Mixed-Use, at a sixty percent design development. These models included retail, hotel and parking garage spaces. The structural models were mostly complete; architectural models partially complete and Mechanical systems less so.
The first study undertaken was the takeoff of curtain wall panels with a sinuous shape where it was problematic to takeoff areas only with orthagonal pdf markup tools.
We had much more accurate results using Assemble to the point where after doing some accuracy checks, we were able to rely on the model extracted quantities more than the previous, conventional methods. Accuracy was improved by 8% on Area quantity extractions.
Our next study dealt with structural concrete volumes, comparing model iterations. Assemble , without any setup, automatically compare two versions of the model, outputting to excel:
- If parameters aren’t in the model to begin with to be able to extract, no 5D tool will work properly.
- Having a medium to export numbers to Excel is an imperative need for cost planners. Manually entering numbers onto a spreadsheet is a timely process. The exporting component is the most appealing aspect of the software for preparing estimates.
- In order to become proficient at sorting the data, training is involved, to understand how data is extracted.
Filtering is a simple process, requiring knowledge of the basics of database search commands
Time savings realized comparing OST/Conventional with revit/Assemble (right hand column in hours)
|Type||Trade Data extracted||Conventional (OST)||Revit||Assemble|
|Walls||Curtainwall||4 hrs but found to have errors as surfaces were not orthagonal||.5 (30 min)||(5 min) 6 X faster|
|Walls I||Typical floor walls-types not defined on the plans, Interior glassdiscovered by viewing model||.75 (45min)||(5 min) 9X faster and more accurate|
|Structure||Concrete Column and Frame study, breaking it down per column mix||3 Day conventional effort||2 HR||15 min 8X faster|
|Steel Structure at podium||Takeoff/Calc Steel wt 4 HR||1 HR||10 min 6X faster|
|Masonry Walls||Takeoff walls OST 4 HR||30 min||5 min 6X faster|
|LT Fixtures||30 HR for complete takeoff|
|Floor Area||Verify Gross areas 1 HR||30 Min||10 Min 3 X|
|Wall /inventory||OST inacccurate 6 HR||1 HR||5 min 6X|
Time savings when extracting quantities for Pilot Projects as recorded doing the extractions:
As a comparison tool Assemble was found useful to quickly assimilate what was changed between different models, especially when in comparison to the amount formatting time required to extract directly from revit. It is a tool for integration managers to use to prepare extractions lists in the format that cost planners desire. The training was minimal, the most difficult task was learning how to filter items.
Assemble Systems allows those without access to the model to log onto a web interface and quickly obtain what they need in Excel. It creates an easier method to sort what is in the model by floor level and by construction category. There is no other application we have tested that can do this functionality, this quickly.
 Applications reviewed: Assemble, D profiler, Exactal’s Cost X, I-construct, Innovaya, Rib Itwo , Solibri, Skybim, Vico’s Constructor and You BIM
 Duplicate items are often caused in design when design options are used in modeling.
 Freedom, by Autodesk, is a viewing software which allows measuring and markups of 3D Individual models and composite (grouped models)
Data that was found easy to extract (Model adjustments listed in italics)
- Concrete- Foundation volume and SF (slabs and shear walls)
- Concrete Slab Perimeter- Slab edges – Family, type, length, volume, count
- Casework – Family, width, count
- Countertops – Family, width and count, custom parameters needed to improve accuracy
- Door Frames – Frame type, material, count
- Door Hardware – Family, type, (count with parameter added)
- Doors – Family, type, fire rating, finish, count
- Glazing and curtainwall panels – Family, type, count, area (Empty panel required removal)
- Windows – Family, type, mark, height, width, area, count
- Ceilings – Type, area
- Duct fittings – Family, type, size, count
- Duct – Family, size, length, count (much of the branch work for typical units downstream not modeled)
- Mech equipment – Level, Mark, family, type, count, model, electrical power, horse power, CFM
- Electrical equipment – Family, type, count, Phasing, amperage, location
- Light fixtures – Family, type, mark, count per floor
- Plumbing fixture schedule – Family, type, trap, drain, count, system, vent connection, waste connection, count per floor
- Railing/ Fence – Type, length(LF), count
- Rooms – Number, name, floor finish, base finish, wall finish, area, perimeter, level (mostly incomplete in 60%DD phase)
- Room types – Name, count, Area per room
- Specialty equipment – Level, Family, type, mark, count
- Stairs – Type, width, number of risers, base level, top level
- Structural column s– Family- Strength and or WF size, level, count, volume (CY), length
- Structural framing – Level, WF, Weight and or size, type, count, volume, length
- Wall assemblies – Level, type of assembly, height, fire rating, width, length, volume, area