BIM Inside Out: Supporting the Owner's Vision for Smarter Project Outcomes and FM

ANDREW ARNOLD: Bruce and I read it all. My name's Andrew Arnold, and I'm here today with Bruce Mace. We'll give you a little overview about ourselves in a moment. The way things will work today--

AUDIENCE: Turn the volume up.

ANDREW ARNOLD: Turn the volume up. If I spoke louder, did that help? OK. I'm going to give you just a little introduction to my company, DPR, and our group within DPR, the VueOPS Group. And then we're going to roll right into Bruce's presentation as a building owner who's implementing BIM for FM at the University of California at San Francisco. And then we'll roll into from there how does a builder support the owners goals and objectives? A little bit of observations on our part and things we're doing. I work at DPR Construction, and I'm involved with our consulting group, which is shifting to this new name called VueOPS. I will give you a little bit more overview about that in a moment. Bruce? Bruce is-- you want to speak? Is your mic on?

BRUCE MACE: Hi, my name's Bruce Mace. [INAUDIBLE]

ANDREW ARNOLD: You want to turn on your mic.

BRUCE MACE: There. Much better. All right. Director of facilities for UCSF Health in San Francisco. And we'll get into it here in a moment. But I come from a construction, programming, and development side my whole career. 12 years ago, I came over to, at the time, what I thought was the dark side in facilities and operations. I had expertise in that, and when I came over to that side of the fence, it was a real eye opener, and that's sort of what this journey is all about to where we're at today.

ANDREW ARNOLD: So just a little bit about DPR. We call ourselves a technical builder. Our markets are advanced technology, data centers, corporate offices, higher education, health care and life sciences, biopharmaceutical manufacturing. Our footprint is really nationwide at this point. We pride ourselves on integrated project delivery types of projects, doing high performing buildings. We have four net-zero offices. Our own offices are net-zero. And we also do a lot of self-perform work. So we do all the way from the very big billion dollar plus projects all the way down to small projects that our special services group handles.

The VueOPS group itself-- I really want to talk with-- kind of work our way into our mission by describing the framework for integrated project delivery that was just published in a book that my colleagues and Martin Fischer just published. And this is-- you look at this as a pull plan, so going from left to right. If you want a high-performing building, you need integrated building systems. And if you want integrated building systems, you need integrated processes. And if you want integrated processes, you need an integrated organization. And if you want an integrated organization, you need integrated information. That's the pull plan. And that's the basis of the book. It's mostly focused on project delivery. It's a really great book. I recommend you buy a copy and read it.

But from the facility owner's perspective, we kind of want to move the goalposts a little bit. Move the chains and look into the operations and management life cycle of the facility, and think not in terms of project delivery data, which is really the model, the design information and the construction information, that cost and schedule. And extend that to looking at, ooh, the trend data coming from the building management system, turning that into analytics about the performance of equipment in the building, service data coming from the work order system, simulation of the building in terms of optimizing performance, and again discovering what's the role of visualization for the facility owner? So these are the things that owners are thinking about now. And our group--

And the reason why owners care about this is because, as you know, those-- by the way, how many in the room are owners? Awesome. Really good representation today. The owners know this very well. If design is a nominal one x, the cost of running the facility for the next 20 to 30 to 50 years, is up to 100 x of the cost of design. So there's a big opportunity there to capture the value that designers and builders have been capturing. Take that value that designers and builders have been creating and doing building information modeling and developing data about the equipment that can roll into or flow into operational systems. And how does that help the owner? That's what our talk is about today.

So the VueOPS Group's mission statement is to seamlessly develop and incorporate integrated information for buildings and operations management to know their facility inside out on day one of taking a building over. Two, meet and exceed design and operational performance goals. Three, improve life-cycle management. And four, prepare for future information and solutions through the use of all this information to help plan capex and opex expenses.

As we have been working with primarily DPR customers, we noticed that they have different needs. They kind of follow a heat map here from some owners, they don't have a model, but they want to visualize their facilities. Maybe they're laser scanning or they're capturing photogrammetry about their building. Some of them want to use BIM models post-construction. Often they're not really clear why, but they're interested. Third, then, there's another kind of category that says, ooh, you're modeling all of my equipment that I care about managing. I want data about that equipment, and I want that to flow into my work order system or communicate to my building management system. And then finally, and this is where Bruce's organization is, off in the deep blue here--

BRUCE MACE: [LAUGHS]

ANDREW ARNOLD: They're interested in enterprise wide implementation, which means developing standards and requirements for all their capital projects so they get a common dataset flowing from each capital project. I'm really, really happy Bruce is here, because their group is really pushing the envelope on what to do and how to do it. So with that, Bruce.

BRUCE MACE: Cool, thank you.

ANDREW ARNOLD: I think I'm going to-- we've got a lot of customers. And now to Bruce.

BRUCE MACE: All right, good morning. Like I said, I came from construction, programming, development, planning, and stepped over the fence about 12 years ago. I came in as the capital projects coordinator for all our capital projects at the UC there in San Francisco, and looking around, quickly realized that, hey, there's probably a lot better we could do. And I had been familiar and trained and raised on IPD, integrated project development, and that's exactly what Andrew had spoken to.

That's BIM and AIA. It's been a guideline. It's been something we've used to compress our milestones, our deliveries. We've reduced cost. We do clash, and then we deliver a project.

My first project at UCSF, 2 o'clock on a Friday afternoon. Everybody's rushing to get a fluoroscopy signed off a room, and I was like, wow, why are we such a rush here? What's going on? And IOR, CDPH, everybody showing up at our doorstep, and they were followed about 15 minutes later by a father and a 10-year-old girl who was getting treatment in the room. And I just saw a big light bulb went off. You know, wow, this is what we do at UC Med. The more we can do with our project deliveries, it's going to help those kids. That girl was the same age as one of my daughters, so it just sort of hit home.

We're going to work backwards. We built a new $1.6 billion hospital, Mission Bay. It came online in February of 2015. It was a partial model. It was built in Revit, AutoCAD. It was mixed model, fabrication models, proprietary, all sort of cobbled together. But we got a 3D out of it, and we had that model. We sort of worked with the project team. It was our first being in the big room with the project team. So we were really involved with understanding how to drive it.

True story, 11:15-- oops, got to go back. 11:15 on a Sunday night, fall of 2016, so still fairly fresh in my mind, got a call. Gone past the engineers, the trade crafts, couldn't find out where this water's coming from. And it was coming out in this chase here. Two hour rated, seven stories tall, and we didn't know what the source was. I got on site same time as my chief engineer, about midnight. We looked at each other, and we ran for the model. So for the first time in my life, we ran for a model.

Got to his desk, and we put an engineer down on the floor here in the cafe, where the water's coming out. Got a guy on the first floor, and we started flying up the model, looking for water systems that impinged on that floor. Two hour rating means it's about two inches of sheet rock, and before you start punching holes in that stuff, you better be sure you know where you're going, because you can cause yourself a lot of damage real fast. You're racing time. We got up to the fourth floor, which was an OR, identified our water source inside that chase, and it took us about 35 minutes. We isolated it. We contained it. We had the mechanical onsite by 6:00 in the morning. We had it all contained for the repair.

He came in, he said, I built it. It's actually over there. And we said, no, Mark. It's over here. We're sure. We know. We contained it over there, where he said he built it. Punched a hole, and it wasn't there. We said, Mark, over here. This is where it is. We're in the model. We were 16 inches off. We had it repaired and started assembling it. The OR went back into use later in the day, and we opened up at 7:30 in the morning. So a use case from the facility's standpoint.

10 years ago, back in 2008, this was sort of-- this is a rough of my perception of integrated project delivery. We would move to the fence I talked about, where the project would get delivered, a lot of this took place, IPD, out here. I was proposing, or I really believed, we needed to get architecture fully getting closer down that end. Project managers, get them past a year, at least a year, involvement with the project. And then our teams, facilities, we should be involved in driving what that looks like, what's going into our buildings, and what's going into the models, so that we can use it. So this becomes important in a minute here. Project delivery to the fence.

This is who we are. We've got four hospitals-- actually, five now. We have 20 person management team overall sites, centralized. We have about 130 skilled trades and engineers. Three million square feet of just I-occupancy hospitals. We do 24,000 service calls a year. And 53,000 pieces of equipment that we need to track and take care of for important reasons.

About 2010, I got to be director, and that was nice. I came on board, and in the first month, they told me, your computerized maintenance system is going away. You've got to come up with a new one and have it implemented in two years. Well, I'm a facilities guy. I'm a builder. I'm a facilities guy. I'm not a let's deliver huge platform IT software.

So we sort of got going on it. We decided we'll put together our own team. We're working with the campus. We got about a five person team. And we bought IBM Maximo after a lot of looking around, because the potentials for what it could do and the promises for a future of integrating with building model. We went to it.

So we went live on our Mission Bay hospital. We had to backfill all of our information, build all the engineering in, all the equipment information, serial numbers. All that it takes to run a CMMS system, we had to load it in there. And it was blood, and it was sweat, and it was tears. It just further cemented the idea that there's got to be another way to do this.

We share our architecture with campus and facilities. We're on the health side. We decided we'd put together a small five person team. That's how big it is right now. And we would drive both Maximo for both campus and medical center. And then we would be the ones involved with the Revit part of it, which we're getting to here. We do that through a governance committee. And we chair that.

Real simple, this is sort of a 20 person management team here. But this is the group that's real important. So I've got a electronic technology manager, a developer/supervisor, and three developers, and they're sort of database scientist developers. We embedded them right in facilities, recognizing that we couldn't really put them out into IT and go through those time periods and the waiting. So we built them right in. We manage Maximo for both sides, both entities, as well as our Revit output.

There we go. This is what really scares you. My kids ask me, what do you do, Dad? I say, well, when I get to a building, I get to our hospital, this is-- that's how I see it. When I walk up there, it's a bunch of systems. If this is MEP, and it's gotten so complex that you really have to focus on what do you really need. Do we need everything? Or do we need a certain part? And focus really well on that. We sort of believe in the latter. If you try to grasp everything, it's going to be probably a bit much.

In 2014, we worked with DPR, EcoDomus, IBM, ourselves, Autodesk, and we sat at a table and we did a no-cost model. We wanted to prove it out that we could do this. We took our energy center off the Mission Bay Hospital and we built it into 3D, completely into Revit. And then we actually took it into Maximo. And then we wanted to view it, were viewing it with a Navis viewer. We just wanted to prove out the concept before we started down that road.

We presented that in 2015, I think, here. These were some of the things that we were really working on doing there. We wanted-- this part right here is real important. Conjecture is that hey, if we could get 5% of that building, and we could specify it, and we could name it and provide nomenclature and a data dictionary-- if we could get that information just the way we wanted and pull it into Maximo, well, guess what? We'd be running our buildings with a 500% better idea of what's actually going on in there. Real important. So the other 95%? It's still there. It hasn't gone anywhere, but we're focusing on, and that's just an estimate, about a 5% model.

I'd like to say thank you to the project team. I know some of you are in the room. R&S, Stantec, Southland, Cupertino, Vantage, Silverman & Light. And then our affiliated partners, DPR, Autodesk, and IBM.

The big news was we came out of Mission Bay. We're all excited, and we had a next building, and they said we were about three years away. So I was getting ready for getting this project going, three years, you know, breathing easy. I'm thinking, this is good. We'll have some time.

Then UCSF hired Alan Ashworth from London, a preeminent cancer researcher. Brought him over. And part of the deal was, hey, you get your own building, and we'll have it completed and operable in three years. So my schedule for delivering a seven story, 125,000 square foot precision medicine building went from three years out to we're starting now. Get on board.

I went to our lead architect, Stuart Eckblad. And I said, hey, Stuart, I want to do it. This is the time. Our team-- we have everybody. We could actually do a Revit build. Let's do it in Revit 2017. Maximo, we're at 7.5. We're going to upgrade to 7.6, and we think we can blend the two. Can we do it on this project?

And he's been around, knows his stuff really, really well. And he said, sure, Bruce, go ahead. Can't cost me any money. You can't bump my schedule. Can't upset the project team. And the contractors can't even know that you're doing it. And I said, no problem, Stuart, that's exactly how we're going to do this. Let's get going.

We're in the big room. We start building. We're in Revit. We're using 20-- whoops. There goes that button again. We're using 2017. It's a fabrication platform. All the subs. We wrote a 58-page BIM execution guide that got baked into the contract, so we had teeth. Real importantly, started working with all our contractors and our subs to say, we want to do this collaboratively. We're not busting chops here. We have a contract, but let's work together to build this bright and shiny object.

So we're in there. We're building things like no-fly zones. Every valve has a 24-inch clear transparent box, deck to deck, that hangs on every single valve, every fire smoke damper, so that my guys can get up there at the end of the project. Nobody can put anything in there. The building's chock full of all those right now, and it's running great.

So we're doing this. We're at 100% CD right now. And here you go, this is in Revit. So this is what a detailer will see on MEP fabrication. You guys all recognize this part here. That's the building. It's getting built in virtuality right now. We're looking at it. We're interacting with it. We're pulling in submittals now. Focusing on a bag in, bag out filter bank there.

Just picking a piece of equipment asset, all of our COBie data, all of our specification, everything we wrote into our data dictionary, our BIM execution plan, we're starting to see that populate out of the model. Goes to COBie and into Maximo. So this is still-- this is Revit. This is what the detailers are seeing as they build it.

We go to the big room about three weeks ago, because we wanted to really sit down. Again, we're always working with them. And we'd had some clashes with, say, somebody putting in electrical conduit, a half inch. Modeling it, that's a piece of work. We said, well, we don't really need that. Show me where my home run is. Show me where I pop out of the wall. Give me my circuits. Let's make this real for what facilities is going to do with it.

So this is the same information, same piece of equipment, and we're going to be explaining this on Thursday. We got another one of these at 3:45. And we're going to show the model. We're pulling-- these are just traps, screen traps-- pulling the same piece of equipment out into Maximo, and we're using the Forge large model viewer from within Maximo. So both of these screenshots here, as you see, those are both Maximo.

We're now in the model. We're pulling in submittals. We're building job plans. We're about a third of the way through with all our job plans to go into our CMMS system when we come live. And that's going to be April of 2019.

Some other folks are taking interest. University of California Office of the President came over and said, we want to trap all this data. We said, great. Come on over. We're doing this. He looked at this and said, so, you can go out in the field and validate this right now? And I said, no, sir, I can't. It's a dirt lot. That's how far ahead we are. This was taken a week ago. Model's 100%. We're building our CMMS, and that's what the lot looks like in construction right now. So that's how far ahead we're getting. So the real time benefits of this approach.

BIM execution plan, really important. Data dictionary, COBie, Revit, all your specification. We're utilizing 360 Glue & Field. Invicara, BIM Assure rule checking. So we're querying the model to make sure that we are good on our information. Autodesk Docs. We've moved around, and we're sitting right here, right now, on how we're exchanging information, doing change management. IBM Maximo. And that's sort of our world. How we've built this model. Real important--

ANDREW ARNOLD: Go back!

AUDIENCE: [LAUGHTER]

BRUCE MACE: We're all taking pictures.

ANDREW ARNOLD: There you go.

BRUCE MACE: Whoops! Oh! [LAUGHS] We've got to get some 3D glasses or something next time. We'll share this, too. It's available, right? This whole deck.

Really important, and this is going to wrap it up here for me. Working with DPR-- where are we going? How are we doing it? There's a lot of pieces. My boss looks at me sometimes and says, that's great. It's getting done. Hasn't seen this part of it, really. It gets really convoluted. There's a lot of work in there and a lot of thinking and mapping out where we're going over the 3, 6, 12 month, and five year sort of--

ANDREW ARNOLD: I think an important point about that last system integration roadmap slide is that-- look at all the tools, and there's different versions of tools. And that doesn't even show the accounts in which all the information sits that the project team-- from which the project team is managing the information.

So what this kind of a roadmap does is it helps an owner figure out what their investments are and what their decisions are going to be when they pull the trigger on making investments to implement technologies and have that technology migrate towards their goal for how they get integrated information access across their systems.

BRUCE MACE: And if we weren't doing this kind of mapping, that's my kind of mapping. This is where we're sitting in our room. We paint the wall. We do all our drawing. You'll see our four solutions that we're striving for here. BIM execution plan. And to this project, we're not going to drop down to here. We're building it right in, and as of Friday, I meet with a new team for our next project. We just keep moving up there.

The real important thing here was in November when this was drawn, this historical fence, I have it right here, and we're driving the specification and the BIM execution plan, and how we're delivering the project from the end there, from where facility sits from an operational standpoint. And the big learn was that that fence is now back here. We're tipping over our architecture, our design and construction, our operational facilities groups into more of a continuum. And I think that's the big corner that we're now turning as an industry. It's working with folks like DPR and Autodesk and IBM that's allowing us to do so.

And with that, I'll hand it back over to Andrew. Thank you.

ANDREW ARNOLD: Thanks, Bruce, for sharing your journey, really. And that's one we've taken with you, since, by the way, DPR built Mission Bay. So we were deeply involved in the journey that he was just describing.

It was out of that and out of working with a couple of other customers, that we're identifying lots of gaps between contracts, division 01 specs, BIM guides, et cetera, that are specified at the outset of a project, and then what the owners are starting to ask for towards the end of a project. So as a builder, we need to pay a lot of attention to that.

So some of these gaps include-- anybody read a division 01 spec recently that specifies turnover requirements? It's great Sunday night reading before you go to bed.

AUDIENCE: [LAUGHTER]

ANDREW ARNOLD: It's very 1972. You know, typically it talks about paper binders, and how you organize tabs on paper binders. Hopefully we're moving past paper. If some of us get to keep marching down this road, we want to move past digital files, too.

We want to move to data and, actually, models of the equipment that-- and I mean computational models, not just the geometry, but models of the equipment that owners care about managing. They're providing in a computer interpretable way the sound, the vibration, the heat, the power, everything you can sense about that piece of equipment as its baseline and during commissioning, so that the owner could then use it for analytics during operations and management. That's the holy grail. That's the mission. We're just not there yet.

BIM guides, if you look at BIM guides on the part of design and construct teams, typically they don't-- they support design and construction real well, but they don't support facilities operations and management so well.

Then another big thing we run into is owners just don't have budget to deal with all of this. Particularly when it's the group receiving, the facilities engineering group, is not at the table with the capital projects group, when we're early on in project delivery.

That's, I think, one of the amazing stories that comes out of the UCSF Health experience is they won the right to be at the table early in the game. In fact, if you look at the requirements they're putting into their BIM guide now, they're driving the bus in terms of owner requirements. And we'll get into that a little bit deeper.

People are still struggling with how are you going to use the information post-construction? And a really big piece of that is change management. At Mission Bay, before we had come to substantial completion of the project, they'd already designed a phase three. And subsequent to building phase three from the Mission Bay Hospital, they've had over 200 changes in the building in the last--

BRUCE MACE: Two years.

ANDREW ARNOLD: Two years. So change management is a big deal. Again, we're working together to develop scalable ways for an owner to handle that and make decisions about when do they change the model? When do they update Maximo? Et cetera.

And then finally, there's a big element here, and I'll hit on this in another slide, which is if you are only thinking about the technical side, the technical details, the technical processes, you're not likely to succeed. There's a very big element of organizational development and communications and outreach to existing stakeholder groups in the owner-organization. And if you don't pay attention to that, if you don't have a plan around that, if you don't have processes around that, chances are you'll deliver all this information at the doorstep of the owner, and it won't get used. And if it doesn't get used, it gets stale really fast. So organizational development is a big deal.

One thing I think builders are learning, one lesson that builders are learning, is the way we think about a BIM for design and construction is very different than the way the owner thinks about it after construction. The owner is really concerned about situations where they have to go fix something. So we call that situational information. The water line leak was a very good example of a situation. And the owner doesn't want to have to navigate through the full design and construction BIM to get to the information they need. They want to go building, floor, room, asset, boom. I got the information I need, and I don't have to munge around in a large model to find it.

For the project team, or get particularly for the builder, these gaps create risk. One, typically for a builder, when we walk away from a project, turnover is the last thing we do with the owner. If there's a bad taste in the owner's mouth about the turnover, it's bad for the builder.

Number two, there's cost-risk as well. I've already pointed out that there's this gap between contracts, division 01 specs, and BIM guides, and what's being asked for at the end of the project. And there's no budget for it, typically. So then it gets you into a messy situation of negotiating extra funding for the cost of annotating all this information into the BIM model and getting it into the work order system. So you really want to think about it at the beginning and address that risk.

Just one more slide on situational information. What we really mean by that-- let's say, for example, there's a hot room call in an office building, in a conference room. That's the real conference room. This is our office. You want to be able to navigate right into the room and get information about the temperature in the room or any other information that's pertinent to that room. And then if you were to go and actually start checking out the thermostat and the fan that the thermostat controls, you want information that's contextual just to those devices. You want to get to that information quickly. You don't want to have to munge around and search for it. We're working hard on that.

So the owners should care about this. They're really-- the business argument is these are the benefits that an owner can get out of making these investments. Number one, extend the life of the capital spend. Get more for your money out of the equipment you're putting in. Meet or exceed building performance goals.

I think Bruce's example of emergency response was very good. We have another hospital that we built in California that had the exact same drain line incidents. They did not have a building information model, and they resulted in over $20 million in total expense in damages.

Save energy. Environmental stewardship. I would say, instead of reduce FTE count, owners are having to manage more facilities with the same FTE account, and that's a real challenge.

So the things we do, our goal is to be an integral part of the building life cycle process, to support the owner past design and construction into operations and maintenance, and to support them in renovation and upgrades.

This is probably a pretty familiar slide to anyone who's been interested in BIM FM. How do you develop the data requirements so that you can work the information into models, get the data out, and then have it flow into computerized maintenance management system. And this is how we do it.

With new owners, owners we're just starting to work with, we typically like to assess the potential for doing it, both technically, their knowledge of systems, the tools they use, what kind of organization they have, how many resources they have, et cetera. What are their business goals? How are they going to get their ROI achieved on doing this?

Then we go into really extending BIM execution plans to go all the way down-- and I'll show this in a moment-- all the way down to the data that needs to be collected. And so let me just kind of fly through this.

Number one, we map the customer type to our approach. We're very sensitive to the needs of the owner.

Two, we'll start off with a workshop. This is kind of the high level outline that we cover in the workshop. Questions we ask our customers about their business interests, their business requirements. What systems do they have? What are their document turnover requirements? Because guess what? Documents aren't going to go away real soon, and there's a lot of content inside of documents that's very useful. So we can't just forget that and go right to the models. What is their plan for change management? Again, system integration roadmap, et cetera. Coming up with key definitions so that we're all on the same page. And encoding all this, really all the data requirements into a product we call a facility data dictionary.

So this is how it works. Number one, this is a kind of an abstraction of what's called a model progression spec. Anyone who's built a BIM execution plan will know what a model progression spec is. It kind of says what systems are going to be modeled, who's going to model those systems. What level of development will they model them at? And that's what that covers.

But we extend that. We extend that to say, for a given system, what's the constituent equipment in that system, and does the owner care about it? And then we go down to the level of development for each piece of constituent equipment.

For each piece of constituent equipment, we then come up with the kind of canonical name for that thing, how it will be named in the maintenance management system, and also how it will be named either in AutoCAD or Revit. It's different. And we're developing standards for how that nomic nature will work. And then we encode it all into this data dictionary. And it fits within this framework of supporting the content authors to have a plan to follow for how they model.

Then running models through a data verification phase before it gets turned over to the owner. At Mission Bay, and partly at PCMB, for the first time I'm aware of, an owner's actually accepting or rejecting a model based on the quality of the data in that model.

We've been working with this company, Invicara, that has a product named BIM Assure, which is a rule-based model checking environment, and it's proven to be very useful on the project. There's other model checking tools out there as well. Autodesk is starting to support some of this type of stuff with their tools right on top of Revit. But it's a very important step. The point is that you can develop acceptance criteria now for the data that's in the models that are being delivered to you as an owner.

The other piece about this is that we don't collect data before we should collect data. Data elements like serial number, install date, and all that, you really don't worry about that until it's installed in the field. You're probably going to catch that during commission test, not during design. So breaking out who's responsible and when they're responsible is a very important piece of this.

In terms of checking data quality, we're checking for the existence of the data. We're checking for the data type, whether it's a number or a piece of text. If there's a pattern to a name, like a room naming pattern, we check for that. If there's a list of approved manufacturers, for example, if there's a pick list, we can validate against. These are the types of rules we can easily right now just check data quality before it flows into your enterprise systems.

This last slide really is important in terms of-- you can do a really bang up, good job on the technology side, but if you don't pay attention to the people, the culture, the communication, tools you have to talk and share what you're doing with everybody in your organization, and your goals, you won't get buy-in. You won't get people to change their behavior. And you're more likely to fail on this side.

So a great deal of our time goes into outreach to project teams, all the way down to the trades, during project delivery, to explain the why of what we're doing, to show them the how of what we're doing, all the way down to training on how to use and configure the tools that they work with to create and deliver this data. What we've found is that the initial response sometimes comes back as a change order or, "That's going to be expensive." But once we've gone through the communication process, explained the why and the how, the what and the when, the costs come down to really no incremental cost to a project. It's very important to do that.

We are also developing a product. It really started with-- it's called VueOPS. It's a VueOPS product. And the goal of VueOPS is to integrate information, both document information and data information, a model of visualization for an owner. And I'm just going to show a couple of screenshots. VueOPS today is a smart, simple, and easy way to find construction turnover documents. So no more going down to the basement and looking for a flat file, or into a bin someplace and looking for a binder.

It supports visualization. We're using the Autodesk Forge environment, embedded into our product, for that. So this is the UCSF Medical Center campuses in VueOPS. This is the Mission Bay Hospital. Diving into that card for Mission Bay. This is the Mission Bay Hospital.

And then this is the document search interface. You have the complete folder tree, or if you wanted a different organization, we can set up a different organization and map documents from a folder. You want to master format instead of the project folder tree, you can do that. You notice there's over 9,000 documents in this document collection. And by typing in a simple search term, like BIBO, for the bag in, bag out filtration system, in one click, you go from 9,000 documents to two. Really fast information finding. And you can open, in this case, it's a PDF file, in a PDF viewer that's part of the product.

Now what I'm about to show you is another one of our sites, and it's some development that we have in progress. It's not yet released in the product. So I'm going to show you a prototype, and I'm going to show you what the final product will look like. We're developing capability to support this ability to filter and slice models by building, level, room, asset really quickly. So what you're looking at here, this is a proof of concept we developed in the Forge environment. This panel here supports slicing. This is slicing level two and level three of this building. You can do multi-level slices. So you can see vertically through the building or just single level slices. And then once you make your slice, you can see the rooms on the level you're looking at. Select a room, and navigate right to that room.

When you click on an asset, in this case, it's a chill beam, we have what we call our-- this is our prototype of the insides panel that lets you see the documents. This is our search engine working in the background against the documents. Also the properties that the facility manager is interested in. It's a very simplified view of all that design and construction information, in a digestible format for the building owner.

What the final insight panel will look like, and it's coming real soon, will look more like this. And it's been designed using Forge extensions as a widget that will be integrated with both work order systems and building management systems. So if you're working in your work order system, in this case it's Maximo, you'd have the viewer for the asset. You'd have access to the documents. And if you were also connected at the same time to build the management system, you'd be taking up the trend data from the work order system and vise versa. It's integrated into the building management system. And if you're in the building management system working, and you want to view the asset from that system, you see the documents again, and you can also pick up all the work orders right there. So this is what we mean by integrated information from the perspective of the facility owner.

So thank you so much. I think we're right there at 9:56. We have a couple of minutes for questions. Are there any questions?

AUDIENCE: So the VueOPS has this [INAUDIBLE]

ANDREW ARNOLD: The question is, can you go back and forth between VueOPS and your CMS or your IWMS? And the answer is, VueOPS will be an add-on to those systems. It'll sit right within those systems you. Won't have to leave those environments. Yes?

AUDIENCE: One of the slides had [INAUDIBLE]

ANDREW ARNOLD: Yes. So the question is, do we have a standard for LOD across the building? Is that-- characterize your question? And the answer is, we care more about assets that the owner wants to manage. So we do go down to LOD 500 for those assets. But not everything in the building. Let's get back to Bruce's 5%.

BRUCE MACE: So when I talked to Stuart Eckblad, I said, I won't rock the boat, and I'm not going to cost you money. We went in, and we got our 5%. We picked out all those assets. And we specified our LOD, 350, 450, 500, whatever our requirement was. All the rest, we left alone and said, that LOD can be whatever it takes you to construct and deliver the building to us. That way we're not layering on added requirements that are going to cost money.

AUDIENCE: So do you-- the Forge integration [INAUDIBLE]

ANDREW ARNOLD: We've invested in our own software development. We're using Autodesk platform in our product.

AUDIENCE: [INAUDIBLE]

ANDREW ARNOLD: We don't have-- that's a really good question. So the question is-- our talk is being recorded, so I'm supposed to repeat it back. Do we have metrics? Do we have measurable results? To paraphrase your question. And the answer is, I think UCSF does. And he's happy to report about that, but not generally across all our clients.

BRUCE MACE: For me to keep marching forward and asking for resources and tools, I need to have business case. So we're just starting to put together business models that I can carry across the street to the c-suite, and say here we go. Last year, we spent $42,000 to measure our floorspace for our licensing. It took a consulting firm to come in and do it, and it cost us 42 grand. Now I'm pulling that information straight out of the model. I did it in 20 minutes, and it didn't cost you 42-- we're putting together those business cases.

ANDREW ARNOLD: If I understand correctly--

BRUCE MACE: Critical.

ANDREW ARNOLD: --you also have really developed some pretty powerful KPIs around your PMs.

BRUCE MACE: Yeah, how we're getting things done. How efficient we are.

ANDREW ARNOLD: And good PMs are dependent upon having a good inventory of the equipment in your facilities in the first place. And so the ability to get a measurable program for your PMs depends upon knowing having databased what's in your building. So that's that.

AUDIENCE: Two questions. The first one is [INAUDIBLE]

ANDREW ARNOLD: Two questions. Can I answer the tagging?

BRUCE MACE: Yeah.

ANDREW ARNOLD: So the data dictionary is the standard for their tagging. The words I was using was words like names, nomenclature, and property sets. So that's all standardized in the data dictionary. The data dictionary is a dynamic document. It's being incrementally improved all the time. We got it done for the PCMB. It's the basis for the next project. It's the same definitions for names and property nomenclature and legal values. As for VueOPS, presently we're working with DPR customers. It will eventually be a product that we release out into the big bad world, but we're still kind of not at that stage yet.

BRUCE MACE: Really wild, getting down on, just real quick, on our building. This [? PCMB ?] is attached the main hospital. Engineers, we go runnin' up to the roof, or my guys, and I'll go up there, and I go supply fan one, two, three, four. They'll get to the new building, which is just a line, and they'll go one, two-- it's same thing, right? Because it's a different build. First time ever, something as simple as its supply fan one through five, and we're starting the new one at six. Our [? VAVs ?] in the main hospital, they named them all different ways. When I called out VAV, I got three in an entire hospital. Well, they named them 27 different things by different detailer. So those standardizations are hypercritical from our perspective.