Owners will spend $40 billion on architectural and engineering fees in the United States this year. What most people do not know is that owners will spend $70 billion on Quantity Take Offs (QTO), cost estimates and cost proposals. They will track the A&E fees, even try to get them lower, but they will do little or nothing about the hidden QTO/Costing costs, if fact I doubt if they have any idea they are spending all this money. Most often when I give out these statistics architects are surprised, if they are contractors they tend to agree. Architects do QTO and cost estimates in-house or use consultants General and sub contractors are constantly going through the QTO, estimation and bid process, and unfortunately they only get a fraction of all the projects they bid. A bid project will have at a minimum 3 general contractors and scores of sub contractors all counting and pricing the same information, and only one sub-contractor from each category and only one general contractor gets the project. Discussions with manufacturer representatives finds some of them spending more than 50% of their time doing "take-offs" for their suppliers and distributors, they are not doing much "representing" when they are spending this much time counting.

QTO/Costing, while not glamorous, is one of the most important activities performed by those working in the construction industry. I am an architect; we work to get a project, hopefully a great project for a great customer on a great site. We work hard to come up with a design that satisfies the program (cost is part of that program, just as much as the spaces that need to be included). We are proud of the design, the owner is happy, we proceed with the drawings, we spend lots of money on the consultants and our employees to put together a set of contract documents. The project goes to bid, the bids are opened and there are a few options for the results 1) the bids come in under budget (great unless you had a percentage fee, you may need to return some money, 2) the bids come in "right on" this is close to a miracle, 3) the price comes in about 20% too high, the owner wants you to redesign on your "nickel" to get the project within budget, all the nice finishes are instantly eliminated, that is not enough so you spend time redesigning and redrawing, 4) the price comes in 40% or more over budget and the project is abandoned, you are unhappy, so is the owner, and the owners looks for a new architect on their next project.

For the contractors, sub-contractors, suppliers, and building product manufacturers the story is different, you get the drawings and specifications, you spend hours, days or even weeks counting, figuring out the drawings, the construction problems and how you will solve them, getting prices on the materials, equipment and labor and you prepare a bid, unfortunately you only get one out of 4 or 5 you bid. Even when you are doing a design-build project or a negotiated project you still spend the same, if not more time on QTO/Costing.

QTO/Costing has different meanings to different members of the construction team for architects it is an estimate, for the contract it is an exact price. To the architects and engineers they need to be relatively accurate, but not perfect. Ideally they would be able to get QTO/Costing numbers almost real time so they could adjust their design, material selections, or inform the owner that they need more money for the project throughout the design documentation process. Without automation this is impractical, the design is changing constantly and meaningful QTO/Costing can take days or weeks when done manually. Often these QTO/Costing activities are done for the architects and engineers by outside consultants and it is cost prohibitive to make updates on an ongoing regular basis. For the contractors, sub-contractors, suppliers and building product manufactures they need exact QTOs as their profitability, and perhaps even their on-going business viability is determined by the accuracy of their QTO/Costing.

In the early 90's my company ASG had a QTO/Costing product that we did in conjunction with Timberline and RS Means that tied into AutoCAD drawings through our software. This was a very difficult undertaking because the AutoCAD floor plan file did not have all the information needed, the floor plan knew nothing about the ceiling which was in a different AutoCAD file. Also there was not guarantee that the various files were consistent, you could have a window in one file that was missed in another file. The final product gave results, but it required more work on the part of the user then they perceived was worth the effort. We needed BIM.

Now we have BIM, all the information in a single graphical model (or a few coordinated models on massive projects) with data either inside the model or associated with the model. So, we must be on the threshold of accurate, easy to use, and fast QTO/Costing. Yes we are, but it is still a very complex process. We are new in the BIM era and consistency, guidelines, and standards are not in a mature state. There are CSI and Uniformat numbering schemes in the US and many other formats throughout the world. The IFC efforts are working toward consistency in naming and attributes but they are not done and are not widely used. There are object naming guidelines by many, including Reed. There are tens of thousands of people using Revit, they are naming objects as they wish, adding attribute headings and then the attribute information as they desire. There are thousands of databases of costs associated with objects (door and frame) and assemblies (walls, roofs). These private databases are found at contractors, sub-contractors and building product manufacturers (BPM's) offices. Each one is different and inconsistent one to the other. There are commercial databases of unit and assembly costs, such as RS Means (a Reed Company). So the challenge it to tie together the Revit project (full of personal names and attributes) with cost data (with random names and conventions). Not easy, but with BIM it is possible.

There are fundamentally two types of QTO/Costing products, 1) estimates with approximate results and 2) detailed costs with highly accurate results. There are several products on the market for estimates and detailed costing. Reed will introduce two QTO products this year. The remainder of this article will look at the QTO/Costing product designed for the architect, the cost estimate; a latter article will look at the detailed QTO/Costing product.

QTO/Cost Estimate
The cost estimate predicts the final cost of the project. The more accurate the prediction the better. The easiest cost estimate to use is the square foot estimate. RS Means has gathered data on the cost of constructing specific building types (libraries, housing, commercial buildings, etc.). You can look up a building type, select a city and quickly see the range of per square foot costs, pick the one that represents the quality you are looking for, multiply times your projects square footage and you have an estimate. BIM contributes nothing to these estimates, other than the project square footage. These estimates are useful to make sure you are in the right "ballpark' but not good enough for coming up with a cost for your specific project, after all you could pick a standing seam all copper roof, mullion-less curtain walls, and Italian tile floors and your per square foot costs using this approach are ‘out the window'. The designer needs estimated costs for their specific project.

The Reed SmartBIM QTO software application (to be released late March or early April) is a unique cost estimating tool that extracts data from the Revit project and aligns that data with RS Means data to provide very fast and relatively accurate cost estimates. I will take you through the challenge of creating the product and the process we used to create a product that we believe will be of great value. The concept for the SmartBIM QTO product was to return results no matter the level of detail in the Revit project quickly, and to provide better and better results as the quality of the information within the Revit project increases.

Data from the BIM Project
The first challenge is the extraction of meaningful data from the BIM model; this is not easy because the naming of objects and assemblies are inconsistent for multiple reasons. While people are good at sorting out inconsistencies (45 minute rated door, ¾ hour) computers and software are not so good, they like to look for one thing and then line it up with one other thing and tell you a result. The quality of the data coming out of the Revit project has many problems. The architect puts in a window, they can name the object anything they want, they can assign any data to the attributes, and they can even add attributes and assign more data of their choosing. CSI numbers were designed to identify products (wood doors, metal doors); BIM on the other hand looks at objects, not products (flush door) and then assigns the material (wood, plastic, oak, metal) as a property. CSI numbers are of little value. Uniformat (Assembly Codes) are used throughout Revit, in some cases they are hard coded to a specific object type, the problem is that Uniformat codes only become useful for specific identification of an assembly after the 5th letter/number and few assemblies in Revit are tracked to this level of detail, so you know you have a suspended ceiling and little more. You can see that this is not going to be easy. Next we look at the other side of the equation, the RS Means databases.

RS Means
RS Means has a massive collection of cost data organized by Unit Costs (based upon CSI numbers), and Assembly Costs (based upon Uniformat, called Assembly Codes in Revit). On the surface this seems good, a closer look at CSI and Assembly Code numbers uncovers the problem.

RS Means gives very accurate Unit Costs for very specific products, an interior 3'0" x 7'0' birch door with wood frame. Because there is not a CSI number for this product RS Means has extended the CSI scheme with RS Means unique numbers. RS Means may have a cost for birch and oak but not cherry. In Revit we may have a 6'8" cherry door; we will not find that door in RS Means. This is not a problem when working manually for the human can decide if the RS Means result is good enough and if not they can add or subtract to make the cost appropriate. Software is looking for specific matches, no matches equals no results.

When you look at RS Means Assembly Costs the same result is discovered, RS Means gives specific costs for very specific assemblies. The problem is now clear, how do you take exact but randomly defined objects and assemblies in Revit and align them with highly specific (but not exactly matching) objects and assemblies in RS Means.

Reed SmartBIM QTO
The simple answer is you create a software product that makes the associations between Revit and RS Means, eliminates unlikely matches and hopefully gives you a cost estimate in less than a minute. On the Revit side we search names, CSI, Assembly Codes and attributes with our software to come up with the best definition possible for the object or assembly. We then take those results and repeat the process on the RS Means database to come up with a subset of reasonable matches, we then average those matches with logic we have created and use that average cost for each of the objects and assemblies extracted from Revit. The result of this process is a meaningful cost estimate, but an estimate that will be off in some cases. The user has several ways to increase the accuracy of the results, 1) you can better define the object or assembly in the Revit model. (Reed is working on guidelines to help you in this process), 2) you can make a better match from the RS Means database (There is a pop down in the SmartBIM QTO product that brings up the RS Means Assembly database that allows the user to select a new specific line item and update your estimate) 3) you can change the costs for any line item (the software will remember this change so every time you run a new report the cost will be based upon your override), 4) you can add new line items.

We believe we have created a break through product: fast and relatively accurate cost estimates that can be run at any time during the design and contract document creation process. The designer can continually create estimates and see relative changes in the project costs. We understand that the first costs using the SmartBIM QTO may not be close enough but have built in the features that allow you to adjust the estimate line item by line item, to refine the estimate. When you run the SmartBIM QTO application it remembers your adjustments and subsequent runs will continue to get more accurate.

Please let us know if you want to be notified when the SmartBIM QTO is released.