Bringing Old School Estimating to the 21st Century with BIM

MELINDA HOFFMAN: All right. Good afternoon, everybody. We're going to start on time so we can get you guys out of here so you can go party tonight. Who wants to sit-in class until 6 o'clock, right? So welcome to Bringing old school estimating into the 21st century with BIM.

I'm Melinda Hoffman and this is David Nachreiner. We both work for LPCiminelli Construction out of Buffalo, New York. And we're going to talk a little bit about how we've accomplished this process.

And also go into some case studies on projects that we've used some different techniques on to help you guys understand where we were and where we've come and where we still need to get to.

So some of the learning objectives today are going to be understanding the link between BIM and estimating, Revit model data overview and how that can accomplish what you need to do and some organization techniques.

Examples of a couple of viewers that we use for our estimators who are non-Revit people and can't use Revit. We've tried. We tried to train them and it did not work. And then quick tips for preparing models for estimating.

So whether you're receiving a model from an architect or an engineer, or you're creating that model in-house, we have some techniques and tips that we've worked out over the past four and a half years that really help us give our estimators what they need and what they're looking for.

A little bit about me. I've been in AEC the industry over 13 years. I started as an intern architect. Quickly realized that I did not want to be an architect. There's too much paperwork for me. I became a Revit technical specialist and now I'm the Virtual Construction Director for LPCiminelli.

My background was architecture and media studies. And I've been working in construction now for four and a half years. My original goal was to really incorporate BIM and architecture and construction, and have them all meld together in a happy world. We've started to try to accomplish that, and one of our first goals was with estimating.

DAVID NACHREINER: My name is David Nachreiner. Thank everybody for attending our class this afternoon. I have about six to seven years in the construction industry. I started off in some project management overseas managing different humanitarian aid infrastructure projects throughout the Middle East.

Came home. Got a degree in construction management and a associates in drafting and design. Spent some time as a civil site designer for a local real estate company based out of Buffalo specializing in commercial retail development and some hotel.

And then I jumped ship over to LPCiminelli where I'm able to incorporate not only project management techniques that I've learned in the past, but also the estimating portion of it. And applying the software and information that we can use to benefit our company.

MELINDA HOFFMAN: So we're going to take a quick poll here. How many people in the audience use On-Screen Takeoff?

DAVID NACHREINER: Too many. Way too many.

MELINDA HOFFMAN: How about CostX? Anybody? One. Bluebeam for estimating? Capterra? Other? One that wasn't listed. OK, great. So we use On-Screen Takeoff, and I say that because there's always going to be those nuances that just BIM can't handle well or easily.

And our guys use On-Screen Takeoff and also another program called Insight, for the civil estimating quantities. But we wanted to get a feel for where our audience was based on 2-D estimating. OK, so how did we start? Estimators, whoops, before I knock over the whole thing.

That's how we've always done it, right? How many people have heard that saying here? But that's how we've always done it, right? So four and a half years ago I started at LPCiminelli and one of my first tasks was to get the estimators to use models that we were receiving. Seemed easy, but it wasn't. They didn't trust the data, and rightfully so.

Some of the models that we were getting were not clean models. They didn't have very good information in them. So we had to start somewhere. So we began looking at the low hanging fruits, or the things that were easiest to take off and validate.

We had to validate and prove those quantities to make sure that they trusted them. And it was a very slow and steady process. So one of the easiest things and most accurate things to take off from a Revit model is typically the concrete and the steel.

So receiving a steel model from an engineer, we can cut our take off time down dramatically. We can pull tonnage quantities straight out of Revit and apply those to whatever they need. And then they apply some sort of percentage contingency for all of the structural detail components.

Because we're not getting engineered models with plates and connection details. So they have that kind of percentage already figured out. But the main beams we can pull off very quickly. Concrete models, I've become a kind of an expert at parking garages.

We've done a lot of estimating for our GC side on parking garages. If we build this ramp, how many spaces can we get, how many square feet is the garage, how much cubic yards of concrete is it. So we started mocking these up. And it became a very quick process.

Less than two hours we could put together an entire parking garage and pool the quantities out for our estimators. This was something that took them a lot of time. And this process was something that was fairly easy to verify. So taking the tonnage costs off, getting our subcontractors' bids back, we were able to compare those numbers and see how close we were.

Then we began a two-way education process. This was the very slow and steady process. The first thing we needed to understand was what the model or what the estimators needed from us. What kind of information did they want out of the model? What they needed to understand from us, was what we couldn't give them, or what we couldn't give them easily.

Which tasks were time consuming to take off in 2D, was something we wanted to know. What is taking you the longest time to take off? Is that something we can cut down by doing a model? Or by taking a model that we're provided and cleaning it up.

The Revit categories versus UniFormat. Our guys were stuck to using UniFormat. That's what they were used to, that's how it was all organized. How does Revit compare with UniFormat? Well, it uses MasterFormat. So then, are all the components assigned to those categories? Maybe. Maybe not. It depended on what the architect did.

So how do we take the architect's information and apply it into a UniFormat spreadsheet that we're used to seeing? And then a breakdown of the buildings. So first one was understanding what it was they wanted.

So the schedules broken up by subcontract, broken up by area of building. This kind of gives a description of different types of walls. And then in the model, highlight what those walls look like. Understanding what we can't give them.

Site work is one of those things that Revit just doesn't quantify very easily. Doing the excavation and backfills, it would take me a lot of time to build an accurate site inside of Revit, where the InSite product can do it fairly quickly.

So these were things where we had to weigh the pros and the cons. What was worth the time and what wasn't. Other time consuming 2D take offs. Tonnage. It took a lot of time to invest in a 2D take off of steel tonnage. Lots of drawings to look at, try to figure out the sizes and try to vet all that information.

Exterior skin, finishes, PCOs and millwork. And I'll let David--

DAVID NACHREINER: So I was walking around Las Vegas Boulevard this morning and I've never been to Vegas before. So this is a really new experience for me. Sensory overload, buildings, lights, people everywhere, really exciting. And I started thinking about our presentation this afternoon.

And as I'm thinking about it, I'm looking around at all these buildings and there's all these different architectural features on the exterior skin of these buildings. There's masonry, there's cast in place concrete, there's different types of finishes.

I'm sitting there thinking to myself, I'm like how much of a pain would it be for an estimator to take a 2D set of drawings of one of these buildings and try to quickly identify and quantify the shell of one of these structures.

It'd take a lot of time, right? Has anybody else kind of got that thought walking here looking at some of these buildings? I mean look at these soffits, for instance. I mean, stuff like that. These little nuances that we find in these projects that take an exorbitant amount of time.

But if we have a model that's produced by an architect or in-house where we're able to quickly identify the shell of that building and identify different areas that may apply different finishes and price points to our overall estimate.

Finishes are one of those things in a model, if they're modeled in the model, it's great. It's a rarity to get a model that has produced finishes, floor and wall finishes. Usually they're just called out on the schedules are on the details.

I find that if we're creating a model in-house for our estimating or pre-construction efforts, I'll model the finishes myself just so we can schedule them separately.

And PCOs, which we'll get into a little while later, these project change orders that we get from the design team throughout the construction process are sometimes a tedious task to identify and separate out from a set of documents that we get from the design team.

And millwork, obviously, if you have a project that has a lot of casework, millwork in it, sometimes it's easy to identify that in the Revit model. That way we don't miss anything from our project teams.

MELINDA HOFFMAN: Yeah, one of those millwork ones would be health care facilities. Tons of millwork in health care models. And we can quickly and easily take out counts with those. And this is just the typical how they wanted to view the information.

So we got into some discussions. Our estimators are kind of picky guys. They're used to seeing it UniFormat. They're used to seeing it in the building envelope, and we get that. So we wanted to understand what their worksheet looked like and how we could take our estimates, pull them out into this kind of information for them.

And lastly, breaking down the model. So taking this building, which is one and a half million square feet, and estimating it based on the areas that the estimators wanted to see.

So whether it's estimating a small apartment complex with retail on the first floor, or estimating this casino that has a hotel, a podium, a parking garage, casino floor, and amenities, that estimators need to know each section so that they can pull the quantities per square foot out for each particular area.

So the hotel knows what to charge per room. Or the parking garage knows how to make that particular money back.

DAVID NACHREINER: How many people in here that create models or edit models use the filters within Revit? Anybody? OK, a lot of people. Myself, coming from a CAD background, site design, that kind of stuff, and typical just overall 2D documentation, I am a huge fan of layers.

So when I really started getting more involved in Revit, the filters tool was my best friend. I can add colors. I can add a name. I can use the comments field and identifying data of the Revit family itself to apply a parameter. And that's what you're seeing here, a series of filters used to identify overall program space throughout the project.

This same technique was applied at a much smaller scale within these respective programs as the project elaborated. And we'll get into those kind of details a little bit later when we elaborate on this project here.

MELINDA HOFFMAN: So what did we do with four and a half years of information? We gathered all this information. One of the things we did was we applied it to creating Revit templates for estimating. So if we're building a model in-house, we have a template to start with that will push the data out that are estimators can use.

First thing was wall types. Are typical models, you know, the Revit out of the box has just generic walls. Six-inch wall, eight-inch wall. And at the schematic design level, that's OK. But when you're getting into more design development and construction design, you've got to know what those wall types are made up of.

So David took the time to create these templates to identify what was in that wall, so the estimator knew how to take it off. Distinguishing wall soffits, storefronts and curtain walls. All those things needed to be identified with what they were.

Storefront versus curtain wall. Those needed to be called out separately, because they're a separate cost. Roofs, ceilings and floors. One of the most common things we do is get a model that has a floor labeled as a roof or a roof labeled as a floor, because that's the easier way to do the slopes. Or because that's the tool that they used.

Well, when you're quantifying that, it causes a problem. So when we build the models, we want to make sure that all that information is correct. And all of that information is standardized within our company.

We also wanted to set schedules so that if we push the schedules out of Revit into Excel, they had the information they needed, not just the standard default schedule or standard schedule that has too much information in it. They just needed specific information.

And then doors. Was it exterior or interior? Curtain wall or is it a different type of door? Is it an ordered door or a custom door? That's all important information to the estimators, as well.

The next process is receiving models. So how do we vet those models? I'm going to let David talk a little bit about that process and the checklist he created for doing that.

DAVID NACHREINER: So similarly, I would say that 90% of the models that our VDC department works with, are models that we receive from AE firms. Architects create them, engineers create them. Specifically focusing on the architectural and structural trade.

So lot of these times these models aren't perfect and we really don't know how much information we can and cannot utilize and push on to our estimation department. I would say that 60% of these models that we receive at times don't have a lot of information in them.

They're not sustainable models. They were created with the intention to represent the client's interest and intents, but after they left the desk, they're difficult to use. So with this said in this issue, we kind of came up with a solution that was based on AIA standards for architectural documentation drawings.

We took a set of SD documents, looked at them, tried to decide what the standards were for them, and then applied that to our models. So as we were talking about before, just a few key examples. Walls, the function for the walls. If they're an interior or exterior use.

If I create a wall schedule and I'm looking, I want to identify just interior partitions to give my estimators an idea of how much chip wear we have in the project, it's going to throw my numbers off. And it's going to throw their numbers off if an interior wall type was used to model an exterior condition.

It also makes my job a little more difficult to try to prove to these estimators that the models are worth even using. Same thing as we were talking about with floors. If there is an exterior balcony system on a building. It's something we see a lot.

So say you have a third floor condition and on the interior it's a slab on deck. And on that same floor we have balconies that people can go on and use. A lot of times, the majority of the time, those balconies are modeled within the actual overall floor plate of that third floor.

That's a different condition. That balcony could be a different structural system. It could have a specific soffit underneath it. It could have lighting coming out from underneath it, radiant heat.

We're able to identify multiple different quantity take offs with that specific quantity isolated from the overall floor plate of the third floor, for example. It also throws off the program space if we're creating a sheet to represent what program we're using where.

MELINDA HOFFMAN: So we're going to get into the case studies portion of our presentation. We're going to talk about four different projects, all different types of projects. And going to explain how we've used models in our process to do better quantities or to create better estimates.

The first project is the medical office building. This is a 350,000 square foot building. It has two phases. Phase one was the core shell and phase two was a tenant fit out. We received the model from the architect at each design phase, which is not necessarily common. So we were lucky to have that information at our hands.

This project is in Buffalo. It's in the medical corridor. It was originally designed as an eight-story building. And this medical corridor is actually a huge development right now in Buffalo.

They have Children's Hospital, which is being built right behind this building. And right next door is the University of Buffalo's Biomedical School of Science, which we've done work on all three projects. So it's a pretty exciting campus and there's a lot going on.

So for estimating team, they needed to get quantities fast because there were so many other projects that needed information. We had to create a GMP for this project. We were fortunate to be involved very early because we were working with the developer directly.

And the problem with this project was they couldn't make up their mind how big they wanted the building to be. It started out as an eight-story building. It changed to a seven-story building. Then they wanted it to be a six-story building. But then they had a possible client that could come in the seventh floor.

So our estimators are going back and forth trying to figure out, OK, well how do we estimate this building if they can't determine how many floors it's going to be. At the end of the day, they decided to design the foundations for seven stories, but estimated it as a six-story.

So how do you do that? How do you estimate this building easily and understand each floor's particular take offs? Some of these challenges of this building for the estimation part were the bump outs. Every little bump out on here had to be estimated separately because it's not a roof and it's not a floor and it's a different makeup.

So it really confused the guys. And having the model to be able to zoom around and show them exactly what this building looks like, helped them to better understand it for an estimate. The work sets from the architect were incorrect.

So even when we tried to isolate areas, core shell or the circulation spaces, we would highlight the work sets that they had and find out that there might have been a wall that was on the exterior that was on the circulation work side. So we couldn't use that as a reliable source for gathering information.

Another one was the wall functions, like David mentioned, were incorrect. Some of the functions were set to interior when they should have been exterior. A lot of the times I find the shafts for the elevators are set to exterior wall types, just because in Revit it's a CMU wall and they defaulted as an exterior type.

Estimating the floor, whenever walls were modeled from first floor to seventh floor. So an architect doesn't model for estimating, they model for construction documentation. Which means that the walls are modeled as high as they go.

But for estimating and for this particular case, we needed to know what each floor level was as an estimate. Not just for that initial GMP, but as tenants purchased the floors and needed to have the retail square footage, they needed to know what that cost per square foot was for each floor level.

So like we assumed, after the GMP was created, they did come back in. And the tenant agreed to the seventh floor. Because we were able to isolate some of this information, we were able to create that pretty quickly.

The other problem was within days of deciding on the seventh floor, they had to order the steel. So we had to make sure we had the tonnage and the cost for all of that ready to go when they said go. This project was pretty unique in that the architect was willing to work with us.

So as we found information that was incorrect, we would mention it to the architect and they were updating it for me. So for example, wall functions. She would go back in and change the functions of the walls, so that every time she sends me an update now, I don't have to go change it myself.

We were able to, like you can see in the image, isolate the seventh floor and easily take that off for estimators to put into their GMP. They also created separate wall types for us, so that we understood the different types of panels on the exterior of the building.

The structural model was also divided for us so that we could easily take those quantities off and we knew exactly how much steel needed to be ordered for that seventh floor. This is a rarity and this project ended up being successful in the pre-construction phase because of the information we had in the model and the willingness of the architect to work with us.

The next project we'll talk about is a hotel. So this was $135,000 square foot hotel, seven stories. And like every client, they wanted a five star hotel for a three star price. Initially when we budgeted this project, it was $2 million over budget. It's not a big hotel. How do you cut $2 million out of this project?

So it was quite a challenge and we were not provided a model for this. So what we did is, we created a model in-house. One of the first things we were able to see is to value engineer this model.

The windows that were designed originally on this hotel were supposed to be wall to wall, heater to ceiling. Very large windows for a pretty simple hotel. It was also a very costly because they were pretty customized.

What we were able to do in the model was quickly change the sizes of the windows to update our quantities for the exterior shell. So we could tell, A, the windows were x-amount cheaper. And B, it increased the wall quantities by this much. However, it offset it by $250,000.

That exercise took five minutes. Because in Revit, when you have all the same window type and you change the size, it updates everything and the schedule's automatic. The other thing it allowed us to do, was it allowed the estimators to visualize.

But we were also able to show the owner that this value engineering proposition was doable. The image on the right shows the window at the smaller scale. Because you have curtains hanging, you're never going to know the window's six inches smaller because you're not going to see that side of the window anyway.

We were able to render to Panorama, Had anybody rendered to the cloud in here from Revit? So we were able to take the design idea for the value engineering, render it to Panorama, and then send that to the owner and have them review our proposition to cut out a quarter million dollars. Wasn't $2 million, but it was a big start.

So this project for the estimators is a big success in showing them what else we can do other than to just pull basic quantities. We can help them value engineer and get what we need to get our budget to where it needs to be.

DAVID NACHREINER: This is a casino project that we're currently in the process of building. We're about six to seven months into the work. It's a 1.6, plus or minus, million square foot building. We have about 14, 15, excuse me, stories of hotel.

We have two to three stories of gaming floor, restaurants, villas, pool space. We have two floors of back of house service space for the casino and the overall maintenance of the building itself. We have four levels of pre-cast parking.

On the back side of the parking, four levels are daylighted and on the front side of the building, all of them are underground. This project here, if you remember a few slides ago, was colored. We had broken out this into our respective program areas, hotel, podium, parking, [INAUDIBLE].

This model was provided by the architect. We had a structural engineer that was very good at producing really high level detail structural drawings. And also, the architect has been great so far producing accurate models for the architecture side of the house.

MELINDA HOFFMAN: So this project has actually been around for about six years. It's one of those that waited on funding, waited on government, waited on licensing. So it has been at the company longer than I have. But the initial take offs we were able to do with this building were the foundation.

So this was one of the first things we were able to get the estimators to agree to let us help them with. At the end of the day it saved them in one take off alone, it saved them eight hours of work.

Now they were able to vet it to make sure it was correct. And they were really excited about the prospect of being able to take that information off and save that amount of time. It freed their guy up to do the other estimates for this project.

The other thing we were able to do with this project was doing the steel take offs. It didn't just give us the quantities, but it allowed us to kind of figure out what kind of crane placement we would need. So this project's so large, we knew we would need more than one crane. But there was a potential that we might need to tower cranes.

If you know anything about the cost of tower cranes, $100,000 a month plus. So we wanted to really analyze the weights of the steel and where the crane placement could go to lift the most pieces of steel where we would minimize having to have a second tower crane. Or potentially just use a mobile crane for those other picks.

We were able to do that by finding the correct placement for the tower crane, which you can see the white crane in the middle of the hotel. And then we were able to utilize at one point would have three mobile cranes.

Also, the crane on the bottom that you're seeing, we were able to size the correct tonnage because there are some pretty big trusses over the casino area that a normal crane wouldn't do and the site restricted us to be able to get the whole way around.

So by taking the quantities of that steel, it was more than just estimating for numbers. It was understanding logistics, which also saved that time and money later.

DAVID NACHREINER: These trusses at the center are a good example of how we were able to use the model to identify quantities. The trusses, there's two trusses that support this two-story condition here. It's basically their convention center. They are 170 feet long by about 40 feet deep. And they're made up of all W members.

Throughout the course of the design phases, from SD to DD to CD, the sizes of the members that made up these trusses changed constantly. And we were able to identify the changes and the adds and change in counts as we went along.

So a program that we use at LPCiminelli that helps us publish models, organize models, schedule models, is Assemble. Does anybody use Assemble at their firms? OK, got a few people.

We find it's very helpful and we're able to allow people who don't have a Revit background or an Autodesk product background to be able to view the model, create their own schedules, create their own views, and look for specific things that pertain to the things that they're taking off.

One of the things that we do at LPCiminelli before we even publish models to assemble, is we edit them in Revit. So going back to those overall program areas, the hotel, the podium, the parking garage and the [INAUDIBLE], we wanted to identify those areas as specific cost centers so we can apply quantities and numbers to those specific areas.

So within Revit we created a text parameter and assigned those program areas using that filter tool we were talking about earlier. Has anybody ever used or know about the Revit database link? Has anybody ever utilized that? OK.

Basically Revit has a tool that allows you to export the database of a model. And with that being said, we're able to export all the data and parameters that pertain to the specific elements that we apply filters to in the model.

What we use this for is, we produced a overall GMP for this project in July. But with most projects, there was design changes along the way. So we were receiving weekly model updates to our GMP model as the project developed. They would come out as bulletins. So we received about 40 bulletins for these models.

Our estimators had a very difficult time tracking the changes that these bulletins added to the overall GMP cost of their models. With the size of this building being over a million and a half square feet, it literally would take them weeks just to vet one bulletin. And these were coming out weekly.

So what we did in the VDC department is, we took the GMP model that had all those filters and all those parameters, we exported the database. What we then did is we opened up the new model that was issued in a bulletin, exported that database, and merged the two databases together in a program called Microsoft Access.

What they are allowed us to do is open up that new bulletted model and turn on the filters and see what remained the same and what changed. And we were able to then identify and grab those things in the program areas.

What we were allowed to do then was publish the model into Assemble. And using the tools within Assemble, we were able into create specific views based on the program areas that we identified.

This tool has been extremely helpful, not only on this project, but on the majority of the projects that we work on that have design changes along the way during the construction process. I've brought along a printout with me that specifies the outline of this. And we also publish it to the AU website for viewing, as well.

Some of the savings that we received doing this specifically with the steel. A lot of the steel changed as the project developed . We were able to identify about $2 million worth of steel that was added from GMP to about present day now, bulletin 40.

This is a manufacturing plant that we're currently finishing up in Buffalo, New York. It used to be an old steel manufacturing site. And now we've turned it into a $1.4 million slab, excuse me, 1.4 million square foot slab on grade manufacturing facility.

The MEP systems in this building are large enough that they could support small towns. The pipes in here are massive. We had a lot of different issues, specifically with structural steel supporting these systems.

So when this building was originally designed, there was actually two structural engineers. One that focused on the structural design of the overall superstructure of the building. And one that designed the supplemental steel to support the MEP systems in here.

Just for an example, of some of the pipes in here were 60-plus inches in diameter. And the duct work was just massive. A lot of additional steel was needed to support these systems, but at the time of the initial design it wasn't taken into consideration how much supplemental would be added.

And how much the steel would be affected once these pipes were filled with water or fluid or whatever was going to be going through them. The worry of the deflection of the steel became a huge concern and they started increasing not only the size of the members of the process steel, but also the amount of the members of the process steel.

These changes were issued by the engineer as PCOs. And every time a new PCO came out they would update their model, which was really helpful. That way we didn't have to assume or apply square footage costs to these areas that they were identifying.

So we were able to take those areas and using the filters again, apply a filter to all the areas that steel was added to to identify the new area so our estimator could vet tonnage and quantities. One thing that this helped us do as well is vet our subcontractors' numbers.

So we had a supplemental steel subcontractor that worked on all this white process steel in here. And as these PCOs came out, he would take the drawings and apply his change orders and quantities to them.

One thing that I've found in the past at past organizations and companies is that construction managers subcontract the relationship. You don't want the sub to come back and try to trick you with too high of costs or too high of numbers. You want to make sure you're vetting every single part and piece that he's estimating for.

But we actually were able to use this situation as a positive exercise. Where we brought the subcontractor in and we said, hey, we're able to identify these areas, these pieces of steel that were added to the model, so let's work together. Let's come up with a number that makes sense and that's accurate.

And we forged a pretty decent relationship with our subcontractor that way. We were able to not only use the filters in Revit to identify the steel tonnage and produce schedules that way, but we were also able to use that Revit DB link.

So as these PCOs came out, and there was probably about 30 to 40 of these that came out, we were able to have an overall federated model that had every single steel add and PCO in it. So that way we knew what was new and what was old.

These PCOs ranged anywhere from just about $5,000 worth of steel adds to some of them were about half a million dollars worth of steel that was added. So this was a humongous effort, not only on the VDC department, But Also our subcontractor, as well.

MELINDA HOFFMAN: One of the things we found out, we did some laser scan tests. It's not really estimating, but I know a lot of you are here in the AC. For this building we did a lot of laser scanning with beam deflection flatness levelness.

David was talking about the large water pipes that were in here. There were four 48-inch water pipes on a single rack with white steel. Before they were even filled with water, they'd already deflected an inch and a half. They had to go back in immediately and fix the steel and reinforce it before they even decided to turn the water on.

DAVID NACHREINER: Another tool that we were able to utilize that was really helpful to us, was actually Navisworks We were able to open up the new Revit models within Navisworks. Highlight the areas that were old. And then highlight the areas that were new by adding just a simple color.

And we were able just to kind of pan over real quickly and identify the areas that were new and old. And that way we could specifically focus on those areas to identify quantities. Like we were talking about before, we utilized the comments field and the filters to apply the actual PCO number to the take off that we were providing.

A lot of, a few times we were working on maybe two to four PCOs at a time, so the schedules we were producing we were able to add the comments field into the schedule itself, and then organize our schedule by the PCO that was being affected.

MELINDA HOFFMAN: When we build our models internally, we actually create a parameter in our default schedules that's called AA underscore Program Area. And it's not because we're alcoholics. It's because AA appears at the top of the schedule. So it just made it easier for us to find.

And then we use that consistently when we model in-house. That allows us to put that program area in there or to use it as an additional field to the comments field. You just do that using shared parameters, if you're familiar with Revit.

In our handout online, it goes to the details of how to create a shared parameter. And some of the other concepts that we've talked about, we did a how-to document or we gave you a link to a great YouTube video. Because why recreate the wheel if there's an amazing video out there that lets you do it?

So at the end of the class, we're asking you to do the surveys that are important to us, are important to AU to make sure that we've done our job. And if you could take the time to go fill that out and let us know what you think, that would be great. And any questions? Yes.

AUDIENCE: What do you recommend doing when you don't have an architect that works with you? [INAUDIBLE]

MELINDA HOFFMAN: So the question was, what do you do when the architect won't work with you on the model. How do you handle getting new models each time? And that's where what David talked about with the DB Link works. So you take the first model, you add in the parameters. You clean up the model the way you want it to.

You use the DB Link tool, which will export all those parameters, all the clean-up you've already done, to a database. Grab the new model and you bring all those parameters back in. Then the only thing you have to do is what's changed. Which is typically a lot less than doing the whole model over again.

Honestly, that was the only architect we've ever had that worked with us that well. Most of them aren't cognisant of what we need. And we are having to go in and clean up models. And this is part of our process, that we know if we want the estimators to be on board and to trust the model, we know that we have clean-up a little bit to do on our end first. Yes.

AUDIENCE: How do you manage to count what [INAUDIBLE]

MELINDA HOFFMAN: The question is, how do you quantify what's not modeled.

DAVID NACHREINER: That's a good question. I come across that issue a lot. So a lot of times I have some estimators that I have to prove that we can use the model and prove what the model can do. In that instance, if the connection is not modeled, then it's not modeled and I can't take it off.

However, it is in the detail drawings. And I will reference the respected estimator to those detailed drawings. We've had issues, we've had projects in the past where there was we were able to use the model to identify not only things that were there, but things that weren't there.

So for example, there was a project that we worked on where there was a pier, there was a column, but there was no footing. But obviously we all know there should be a footing underneath a pier in a column.

So I was able to identify the areas that weren't quite finished yet in the model and reference the specific sheets within the model and they were in the detail drawings. We were able to take them off that way.

So it kind of worked a little backwards in that aspect, where I couldn't produce them a 3D element or a schedule based on a 3D element, but I was able to identify areas in the drawings where there were schedules and detailed drawings for them. Does that answer your question a little better? OK.

Did you have a question?

AUDIENCE: [INAUDIBLE]

MELINDA HOFFMAN: Yes, absolutely. The question was, if you had an architect on board in a design build situation, would you try to set up a template. Absolutely. If they are working with you, you want to make sure that some of those things that we've talked about with the functionality of walls, roofs and where floors end and roofs start, all of that's laid out at the beginning.

Just like when you're starting a project with an architect and a structural engineer and you're talking about who's modeling what, you have that conversation of collaboration.

DAVID NACHREINER: That's a great chance, too, to forge a relationship. I don't know about anybody else in the room, but a lot of the architects we work on on some other projects, we've worked on in the past before.

So if you can forge that relationship and that rapport with an architect, chances are they're going to be an architect in a project you're a CM for again. And it's just a great way to kind of forge ahead with that relationship.

MELINDA HOFFMAN: Does somebody back there have one? Yes I'm going to have to move so I can hear you.

AUDIENCE: [INAUDIBLE]

MELINDA HOFFMAN: Are you using Assemble for your cost estimates, and does it talk to your database?

AUDIENCE: [INAUDIBLE]

MELINDA HOFFMAN: Inserting the cost per units.

DAVID NACHREINER: We don't insert the costs directly into Assemble. The main, and I should've mentioned this earlier. The main thing that, a few main things that we use Assemble for is the model comparison tool where we're able to compare iterations of a model.

So like I was saying before, we export the models right out of Revit with those program areas in them. We're able to use Assemble to create views of say, the hotel. And then we're able to compare the views of the hotel between model iterations. That's probably the main thing we use it for.

We also use it for, so people can actually see and pan around the model a little bit, as well. The actual cost input into the program is not something we've entertained at this time. But we have talked about it in the past.