Transcript: Blockchain and BIM Design Improved with Game Theory and Machine Learning

Blockchain is not only about the big brother and knowing everything about everyone. It's making sure everybody knows about everybody else's activities and by doing that you create trust, you create transparency. But blockchain has also this property of making data more machine readable so you can automate things much better. The advent of blockchain and the advent of property in data of being mutable, the fact that you know where the data came from and you can guarantee that as well. You can calibrate your game theoretical model, you can calibrate your utility function in a integrated team and of course the perfect framework for that is a build information model because if you have the build information model running on a common data environment then you really have the kind of infrastructure to run a game theoretical model to assist the team to make the best decisions they can make.

One thing that we don't do very well in the construction industry is value the data we create. Often the construction industry is complex and fragmented and there is a lot of variation and fragmentation. I think we need to think about the data we are creating. It's very important. So looking back, look at what you've done, what your practice has done, see how you can make use of the data you create so it benefits the next cycle.

Hello, I'm Brittany Campbell Turner and this is the Constructor Podcast, episode number 48. Hello and welcome to the Constructor Podcast, the best way to build it. This podcast is dedicated to helping property owners have certainty in their decisions about their construction projects. We talk about fostering trusting relationships, helping you to understand how to lower risk, be under budget and on schedule, but most importantly, exceed your end user's desires. Last week, Jeremy Barnett, one of the founders and a director of the Construction Blockchain Consortium and a practicing lawyer, spoke with us about how he became interested in blockchain.

He had been doing research on algorithmic dispute avoidance and how that can improve the process of a historical evidence and record review. He also spoke about smart contracts and how they can be implemented from a legal perspective. If you missed this episode, check it out at constructor.com slash ep47. Today is the fifth episode of the five part series where I interview guests about the use of blockchain in construction. Today's guest, Abel Maciel, is an architect and researcher with specific expertise in BIM.

He is also the main coordinator for the Construction Blockchain Consortium with responsibility on the steering committee, secretary, and heads the technical committee. We talk with Abel about how BIM design decisions can be impacted by game theory and how game theory would work hand in hand with blockchain to improve design decision making and knowledge transfer. We talk about how we need to start looking at the data we're creating on projects to maximize the value of future design. Before we actually get into the interview, I wanted to give you an introduction to some of the concepts that we'll be discussing.

Just want to give some context about specifically these three areas so you can get a good understanding before we discuss how blockchain and BIM can be applied. So first is behavior economics, second is game theory, and third is machine learning. Behavior economics is the intersection between economics and behavior psychology. It provides a framework to understand when and how people make errors. It is the review of orderly decisions or biases that recur predictively in a particular situation.

Behavioral economics help us to understand why only one third of Americans floss daily, why most people's expensive home treadmills turn it into overpaid coat racks, and why motivating humans is more complicated than ever before. So moving on to the topic of game theory, game theory is the science of strategy. It attempts to determine mathematically and logically the actions that players should take to secure the best outcomes for themselves in a wide array of games. In game theory, there are two types of games. The first game is a zero-sum game.

It is a game in which the gain of one player comes at the expense of another player. The second game is a non-zero-sum game. It is a game where the gain of one player doesn't come at the expense of another player. It can be mutually gained or mutually harmed. It comes up in really three different ways, where you have mutual wins, mutual losses, or conflict.

But the games all share the common feature of interdependence. That is, the outcome for each participant depends on the choices or strategies of all. The last topic that I wanted to talk with you guys about is machine learning. And simply, machine learning is a science of getting computers to act without explicitly being programmed. Speech recognition, web search, and self-driving cars have been born of machine learning.

So all of these three topics we'll be discussing, and we'll be discussing them interchangeably. Abel will go into further explanation on how they interrelate and can be combined to make smart contracts even smarter and improve the data in the building information model to be used on future projects. So let's get into the interview. Hello Abel, welcome to the Constructor Podcast. Hello there.

Thank you for inviting me. You are an architect and researcher with expertise in building information modeling, and currently you're the main quarter at the Construction Blockchain Consortium. So I think that's great that we're going to be able to talk with you about design, particularly in blockchain, and you work on projects with Google. Can you tell us where you have worked on projects with Google? I've been helping Hadwick Studio and also Big, the Archangels Group, on the building information model for the Mountain View Campus.

What has your experience been in relation to collaboration? What would you say the benefits are of BIM? I think the benefits are fantastic. The entire team, Mountain View Campus, is absolutely huge. So we have 20 architects also in New York, about the same amount here in London.

Also a very large team in San Francisco, and the engineers, they are based in LA. So it's a geographically very distributed team that Google made sure that put together the best experts in their fields to work on this project. To exchange information, three-dimensional information and design information with people geographically distributed is very difficult. It's just a very problematic thing to do. With the advent of BIM, the shared common data environment and collaborative processes in BIM, we managed to do this efficiently.

So we can actually see when the engineers upload a new structural model, and we see it as soon as the model is synchronized, which takes about, in a good day, it takes about five minutes. In a bad day, it takes a couple of hours to synchronize the model. But we can see what they are thinking, how they are developing the structure of the building. And then we can work on our side, on the architectural design side, adjust accordingly or contact them, see what's going on and why a certain strategy is being used.

And the same for all the other experts working on the model. So it's really fantastic. And it streamlines the entire information exchange process. I know that BIM building information modeling here in the US maybe considered something a little bit different than what it is considered in the UK. Just because you're absolutely utilizing BIM in the UK right now, could you give us an understanding of what BIM is?

So yeah, there are a few differences between the US and the UK. The UK did quite a lot of work in the last few years in some ways trying to catch up with our neighbor countries like Denmark and Sweden. And now we have a very comprehensive definition of BIM and BIM level 2. So the main document is the PAS 1192-2. BIM is very much a way of working as well as a technology.

So with the advent of building information modeling software, we are able to do much more now with digital prototypes of a building. So we not only model the building in three dimensions in the computer, we can also embed a lot of information. And the BIM protocols and the BIM levels, they try to in some ways dictate a workflow that the integrated team can work together on this federated model to create the construction documentation. So the levels really, what they mean is they are about how close the team is working, how well they are putting the information together and how they are using this information to build the building.

BIM level 2 specifically is interested in this model to be created with a reference with in a federated way as I mentioned before, with the exchange of information with other models. The architect will have his own model, structure engineer will have their own model and so on. They are federated to show the complete design strategy. With BIM level 2, we also have to have a very comprehensive employers information requirements, the EIR document, and this needs to be a clear definition of what the employer requires for the building.

So all the information they will need in the building information model and if they are using the building information model to run the facility management of the building, then they have to highlight this on the EIR and BIM level 2 also deals a lot with suppliers and the supply chain. This information needs to be embedded in the building information model, the BIM objects or if using a software like Revit, the BIM families. To run the entire project, we have to have a document called the BIM execution plan.

Over here the acronym is BEP, I think in the US it's BXP, but at least in California it's BXP. This document really defines how the team is going to collaborate, how often we're going to exchange information, how often we upload the information for this federated model that normally is hosted in the cloud in a common data environment. The common data environment is also a requirement for BIM level 2, so integrated team has to run the building information model, the federated model in a common data environment and share this information with everybody connected to the project.

It is really about sharing information and sharing information in an effective way. There is no abortive work, no problems with people going away and working very hard on a problem to only realize that that problem is no longer a problem or the nature of the problem changed. There is a waste in a professional's time. According to the PAS 1192, this is very much what defines BIM level 2. Normally the output for BIM level 2 tends to be 2D drawings extracted from the 3D model.

When we start to talk about BIM level 3, the idea is that we actually use the 3D model to construct the building. So you no longer will need to print drawings to take the drawings on site to actually build the building and so on. It's going to be more about using things like augmented reality or some sort of projector. I think there are some people using laser projectors to project parts of the BIM on site so you can see what you're constructing and so on. So it's about streamlining the creation and the use of information.

Thank you for clearly laying that out for us. I know on the podcast, we've discussed it a couple of times, but you're utilizing it actively and it definitely changed your mindset about the typical schematic design development. It's a much faster iteration and I like the fact that there is first of all the EIR that establishes what information is decided upon that's going to be actually incorporated into your objects or your families. What's the information going to be used for? Is it going to be passed on to the facilities management group for the operators to utilize?

And then in addition to that, just having the moments in time where you've decided you're going to upload certain type of information to the common data environment. I think it makes a lot of sense to always be on the same page about what information will be shared at what time so that you can make decisions about what you then will react to and have that conversation about it as well. So it sounds like that it allows you to be collaborative and even more innovative. I think it's a new thing.

It's a new way to collaborate. If it's not managed very carefully, it can cause problems as well. The question about how quickly we share information when we're developing design and when we have internal review meetings. We sometimes hold some ideas back because we think that maybe it's something to be first discussed with a particular group of people, maybe just the client and then integrate this to the beam. And when we have this very fast collaboration process and exchange of information, sometimes we have to manage very carefully how we share information.

I think that that's really helpful. So could you tell us how blockchain may impact? There's lots of broad ideas that we don't know the answers to. I think that this is truly relevant when we're thinking about how to apply the utilization of blockchain. Blockchain is really a wonderful piece of technology.

It's a new type of database architecture. It's a distributed, supported architecture for database. What is interesting about blockchain is the properties of immutability and provenance of data. So this is really core to the discussion. And when you can guarantee immutability and provenance of data, you can start to understand what sort of utility people are realizing on that data.

There are many things we don't know and there are many tools that we are imagining today that they don't exist yet. There is a lot of speculation on how blockchain and other intellectual tools like game theory can help on design. Can you describe what game theory is and how that relates to blockchain? Game theory is really like an exogenous decision-making modeling device in economics. So by exogenous, I mean the decisions that the two or more people make when they have to collaborate or when they have to coordinate something.

Opposite to behavior economics, which is more an indogenous decision-making process, game theory is more interested in the information one individual has, how that information will impact on the decisions of some other individual or a group of people. And the opposite as well, how you gather information so you can have a better informed strategy to make the most rational choice. You can possibly make. Underlying all this, we have a utility function. So how important something is for you and how not important something else might be for you.

This utility function informs you the best choice you can make. So it's really about decision-making. The advent of blockchain and the advent of property in data of being mutable, the fact that you know where the data came from and you can guarantee that as well. You can calibrate your game theoretical model. You can calibrate your utility function in a given game theoretical model to make the best choice possible.

And you can really design an interaction method so you attain the best result possible in a collaboration in this integrated team. And of course, the perfect framework for that is a built information model, because if you have the built information model running on a common data environment, then you really kind of infrastructure to run a game theoretical model to assist the team to make the best decisions they can make in any given point in time. I think it is possible to do it without blockchain, but you wouldn't be able to do it very well if you don't have this immutability and provenance of data.

Because of blockchain and because of some scripting language we have out there that uses blockchain, we have this new field of smart contracts. They are actually starting to be very disruptive because they are little robots as a cute team contracts and distribute information to a team or to two or more parties like a buyer and a seller and so on. With game theory, we have the opportunity to make smart contracts even more smarter. I think that helps us to understand how that is utilized with the concept of game theory in BIM.

So it would lay out the context and framework. Like you mentioned, the provenance of data itself, it would lay all of that out very cleanly and understandably in say your smart contract and the code. And that would enable the transactions of information. I'm just trying to make sure I have my head wrapped around it properly. Nowadays we have things like Ethereum and Bitcoin script and things like that.

You can write smart contracts. But when you're talking about the building industry, you're talking about a very complex, very litigious industry. The contracts for construction projects, they are very complex. They are problematic to automate. So it's very easy to automate.

If you want to buy a book from amazon.com, you can run a smart contract to manage this process and automate this process. But if you're talking about a very complex construction project that will take maybe 15 years to be completed and then you have all sorts of data flows and consequential issues from the building. You have the data facilities management problem that comes after that. If you want to automate all that using a smart contract, I think today it's not really feasible. But with this intellectual framework and with the mathematical devices from game theory, you can evolve the smart contract scripts we have out there to accommodate for the complexities and the potential pitfalls you have in the construction industry and construction contracts.

Is there an example that you can kind of give us to help us understand maybe how this would work? It is something that we would dream of actually. It is very complex. Let's imagine a situation where a smart contract could evolve with the initial steps of applying smart contracts to construction is building code compliance and supply chain monitoring and certification of products and equipment arriving on site. Building compliance goes really well with BIM.

So building that has to have disabled access, it is a monumental task. We have lots of very old buildings here and some of them are not very accessible. There is a lot of work to be done on this area. The new buildings are compliant. But when you design a new building, the parametric families in BIM could activate the smart contract to flag some problem with the access of that building.

We do have some software out there that helps with compliance. With a smart contract, this could be much more efficient, I believe. The smart contract would notify the relevant party if there is a problem. So parametric BIM family could flag something to the smart contract and the smart contract would manage this process, this communication problem with the relevant party. But at the moment, you have to do this pretty much semi-automated.

You need to go and check and then the current solutions, you will have to inquire the building information model to find problems. You can create some schema that will help you to find problems in the building information model, but if you don't look, you don't see. With a smart contract, you probably would know if the problem is there because the smart contract itself, using BIM families in the blockchain, would send that notification to the right person and that person would contact the designer, for example, for final clarifications or to ratify the problem.

The other example would be supply chain and I think this will be highly disrupted with blockchain and BIM. I do get lots of emails from people asking me how to better put a BIM library for the products they sell. And this is very interesting because once you have that BIM object, they will behave in a certain way. You know you're designing something that works well with that product or equipment. But with a smart contract, we have three things going on.

You have the building information model itself, which is a representation of the real object. And you have the blockchain and the blockchain could be the nexus between the physical object and the building information model. If there is something on the object that is not the same as what you see in the building information model, today you wouldn't know if the entire manufacturing of that object is blockchain. Then you would know for sure when it's time to install this on site, etc. If something is wrong or if something is not operating the way it should be operating or is not installed the way it should be installed, the smart contract would flag this very quickly.

And of course, this would have also an effect on project insurance because the insurance companies would know exactly what is going on and therefore they can run their business more efficiently. We talked a little bit about supply chain in one of the past podcasts with Jordan Williams with IntelliWave and they're utilizing smart contracts with material management. You can find this episode at constructor.com slash EP 46. So does that mean blockchain helps insurance companies with accuracy and to be more transparent? Yes, yes, I think so.

Thank you very right in saying that. Because once something is on the blockchain, once you have this distributed ledger and you have the blockchain operating and the data is captured in that way, there is no ambiguity. It's very, very clear that the data came from a certain place and the creator of that data was a particular person. It becomes very clear to automate this process and trying to make the process machine readable, then blockchain would be ideal for that. It's not only about the big brother and knowing everything about everyone.

It's making sure everybody knows about everybody else's activities. And by doing that, you create trust, you create transparency, but the blockchain has also this property of making data more machine readable so you can automate things much better. I'm curious as to how typical lead times can be built into your BIM model, into the object, and it can be planned for when incorporating into the design. In addition to that, how can it can be verified on the blockchain that it will meet the timing in the schedule that is built into your BIM model?

Yes, because you do have the dimension of time in BIM and you do have the dimension of time in smart contracts, you could conciliate the two. I mean, there are differences. Like in BIM, you have different ways to define time. You have milestones for construction projects, you have existing and new, and you have different options as well. So you can have the same geometry configured or arranged in different ways.

And there are different representations of time. And you can also have variables in the BIM that they deal with a certification of something being installed or being received on site, etc. So you can conciliate the two, yes. The great thing about blockchain and smart contracts and BIM is you know you're not only entering data on the build information model and that data can be tampered with, can be changed, you can do something to that data if you want. You know that data is representative of what is happening in reality on site.

So this is a very big plus in my opinion, because build information model is doing a good job in terms of making sure people build things that actually will work or they're actually doing what they should be doing. For instance, it's very hard to build a staircase in a build information model that does not work. To have a staircase that the steps are too high or too low or it's not usable in some sort of way, it's very, very hard in BIM. You can do it if you want, but you would have to hack the BIM software quite a lot to design a bad staircase.

There's so much to be done in terms of improving what BIM can do for the actual construction of a new building and to increase quality. And I think part of the answer is in how we're going to utilize blockchain, how smart the smart contracts will be. Yeah. Well, and that leads us kind of back to the original topic, in my mind, at least a transition of utilizing BIM level two to BIM level three, because BIM level three is hopefully utilized to actually construct on site. And you're working from that 3D model and you're pulling from the actual objects that have already been represented in the BIM model, placing them physically in the real space and the actual built environment.

I completely agree. So in the world of BIM level three or and beyond is going to be very much about how to make 3D prototypes, digital prototypes of the building. And we use this term digital prototype because it contains so much more information about the 3D nature of the building, but also everything else that needs to be done to construct that building reality. How to make that digital prototype useful for the builders, for the contractors. Contractors, they don't need to see everything at the same time.

How the BIM is visualized, it will be critical because it is far too complex to show everything at every single time to every single person. So it needs to be filtered and manipulated in different ways to inform the construction team. I think this is very key. So it opens up lots of opportunities for augmented reality, virtual reality, and so on. Because you have to operate in this virtual environment, this will also be the user experience that we will be creating our blockchain database, also certifying things according to what we see on site.

We'll all be capturing the blockchain as well. So it's like a retriving and writing on this distributed ledger by the augmented reality user experience. So this at the moment is in very early days. There is nothing that I believe is the right answer for how to deploy augmented reality on sites. People are using iPads and other tablet computers and phones, smartphones, et cetera.

But to use your tablet computer, you need to use your two hands, or at least one hand. And then you need your hands free to actually do the work, right? We need to rethink. It's an ergonomics problem. It's something that's very needed, but it's at the moment not happening in the right format.

We need to rethink how the user experience is going to be delivered on site to help the construction team to operate the smart contract, to retrieve and write data in the blockchain, how this is working seamlessly with the building formation model. We don't have an answer for that yet. We are in very early days. We have a sense that the technology exists, but it's not integrated well enough to constitute the right answer. I do want to make sure to touch on behavior economics.

This is another topic that I think you can speak very well to. And could you explain to us what that is? We can talk a little bit more about some technology as it relates to that. Behavior economics is almost the opposite of game theory from the perspective of the interface of the decisions. So in game theory, it's this idea that when you're making a decision with somebody else, you need to somehow externalize some information.

That information will be computed by this other agent or other person. This will inform how they want to go about making a choice in relation to that situation. So it's an exogenous model of decision making. With behavior economics, it's a kind of decision making. It's like you yourself make a decision about something happening to you.

So you have nature, you have the environment, and then you need to make a decision. A very simple example is it's raining. Decide if you will use your raincoat or if you will use your umbrella. And it's really up to you to make that choice. So you have the environment, you have the rain, and then you make that choice.

So lots of behavioral models, they use different menus to categorize different decisions or to calibrate the utility of your choices in different ways. So there are lots of risks and prospects, and you have different internal models to make that decision. And it is very connected to machine learning. Machine learning, it's a field in computer science that gives computers the ability to learn without being explicitly programmed. When computer scientists develop some machine learning code, it's not like an expert system in the sense that they write every possible behavior of what an expert in that field would do.

They just write a machine learning code that might be mimicking some biological process like neural networks. The program itself learns how to deal with that situation. And you have different methods of doing that in computer science. Over here in London, at the University of London, people are very interested in support vector machines. But recently, in the last few years, because we have more computational power and we have a better understanding of neuroscience, neural nets are really coming back.

And they are proving to be very powerful. Of course, nowadays we have deep learning, which is basically layering many neural networks and software, or little programs, layering them, processing them as a batch. Before we couldn't do that, now we can, and it's proving to be incredibly powerful. It's also proving to be some sort of generic intelligence. You can put it to learn wherever you want it to learn, and it will adapt very quickly, and it will do things extremely well.

Deep learning nowadays can recognize images better than human beings. That's quite striking. When you add the lay of behavior economics, then there is something really fascinating going on because machine learning is on its infancy as well. So it's developing very quickly, but it's not doing as much as we believe it will do. We believe it will do much more, and eventually it will be something like a person, all the complexities and intricacies of a real person.

And behavior economics gives some indication of what direction we need to go. Behavior economics is also changing quite a lot. There are lots of hybrids between behavior economics and game theory. Lots of economists are writing models of behavior economics, which behave a little bit like a game theoretical model, which is very exciting as a hybrid. It's truly fascinating, actually.

I quite like the concept. I think this gives us a sense of the opportunities. Because there's so many opportunities, and I do want to make sure that we are practical and down to earth when we're looking at what to do next. The podcast, we talk a lot with owners, and yes, obviously designers and contractors are listeners too. But I guess what would you recommend for the constructor community in order for them to be prepared for design innovation?

What kind of mindset should they be embarking on? The construction industry is complex and fragmented and have many different sectors, and people design different buildings in different ways. There is a lot of variation and fragmentation. I think a cultural question like that is about looking back, look at what you've done, what your practice has done, see how you can make use of the data you create so it benefits the next cycle. I think we need to think about the data we are creating.

It's very important. So I think all the things we've been talking about, build information modeling, blockchain, technologies, game theory, machine learning, behavior economics, they are very much about data and how data is manipulated and how to extract value from data. One thing that we don't do very well in the construction industry is value the data we create. Often we generate terabytes of data in a project, really a lot of data. We sense that that data is extremely valuable, but somehow that goes into a service somewhere and we forget about it.

And then it's just later on when we do something similar and some more experienced person in the team or in the practice remembers that type of work has been done before and we could benefit from having a look at that data. It's just then that we realize that we can gain from the data, but often we just carry on and we never look back. I guess what I'm saying is we need to be more careful with the way we produce data and we need to think in a more systemic, more cyclic way.

We should think about how we could recycle all the data we create and how this could benefit what we do. Because if we try to identify or if we succeed in identifying those different pathways to reuse data, we actually write in what we could call the white paper for some machine learning method or for some smart contract method that will help us in the future. I think at the moment we waste a lot of the work we do. We're capturing the work, but we're not unleashing that value, the different technologies we're using at the moment.

We're walking toward this reality, I think this new horizon where we're realizing that we have a box full of gold, sort of, say, in servers or hard drives. We just need to learn how to unleash that value. This is a very important point because we have all this fantastic machine learning methods appearing and they can do incredible things, but they need a lot of data that we need to kind of teach them or at least tell them what we want. It's not simply designing a building, handing over the keys to the client and then never looking back.

We need to think about the whole thing as an ecosystem. This is really key. And if we can do that, then we can speed up the development of the right technologies. We will also take the right steps to be ready when this technology arrives. Wow.

That's an excellent response to that question. And I can't add anything to that. So thank you for that, Abel. Okay. So this has been a real fun time just kind of going through BIM, going through blockchain utilization, behavior economics and machine learning and all of that.

And I think give people the opportunity for people to find you on construction. On construction blockchain.org. Any other ways that people can reach out to you? How can they learn more about what you're doing? Yes.

I spent some of my time at the university college. I'm a senior research fellow there at the Bartlett School of Architecture. I'm on the website there. I have my own website abelmaceo.com. Something that I'm doing in parallel with construction blockchain consortium and my own practice is the designcomputation.org initiative.

This is a design computation wiki and it's open to everyone. We run meetings here in London, but I would like very much to run meetings everywhere. And the idea is to bring computer scientists and designers together and cross fertilize the two fields. I know lots of computer scientists working on wonderful tools that could be definitely used in the construction industry and in design, but it doesn't happen. And also we have lots of fantastic ideas in the design world and in the construction industry.

We need to solve problems in very unconventional ways. And I think the computer science community could learn a lot from design as well. The idea is to create a knowledge platform. And this is what the design wiki is where people write about design and computation. And if you have an interest in be part of that, the address is designcomputation.org.

So yeah, thanks for sharing about what you're doing at designcomputation.org. I'll definitely be putting that in the show notes. So with that, I just want to thank you, Abel. This has been an excellent time spending with you. Thank you very much.

Thank you for the opportunity. Abel and I had a super interesting conversation. So let me know if you enjoyed our discussion by connecting with me on Twitter at Brittany Natchezcora C.T. or find me on LinkedIn. Or you can just email me too at Brittany at Constructor.com.

Again, for the email that's B-R-I-T-T-A-N-I-E at ConstructRR.com. This is the last episode in the series, but believe me, this is not the last of blockchain and construction investigation at Constructor. Blockchain brings transparency to organizations through accurate shared record books of transactions. We cannot get away from that at all, but for long-term behavioral changes, Clarity First comes first. So I wanted to give you guys a recap of the LCI Congress.

And I was able to do a one-on-one interview with keynote speaker Karen Martin and talk with her about Clarity First, her new book that will be released in January 2018. She says that ambiguity prevents organizations from operating with focus, discipline, and engagement. So we talk about how organizations don't have clarity, some of the pitfalls of setting a clear foundation, relying on ambiguous goals, and lastly, some of the recommendations that she has in order to be successful. Organization implementing clarity first. So during the Congress, I was also able to hear from a mix of owner-operators, designers, lean consultants, general contractors, on how they are capturing and leveraging the lean advantage.

If you guys didn't know, that was really the theme of the Congress. And so I asked questions about what they're doing, basically to make impacts in their organizations, on their projects, and in their AEC communities. Last but not least, I've asked them about their perception about the AEC world in the future. So you'll be able to hear what my guests have to say next week. I look forward to talking about the 19th annual LCI Congress Recap starting next week.

So don't forget to subscribe at Constructor.com to get email updates from me. If you haven't already, subscribe to the podcast. You can do so at iTunes, Stitcher, Google Play, and TuneIn, and actually just recently, iHeartRadio. So I'm working on just kind of getting on all the platforms for your listening convenience. Please leave a review to show your support and let me know you're enjoying the podcast.

I have to thank you guys. There have been a lot of people leaving reviews lately, and it's just so encouraging. So I have to thank you here on the podcast. It's been really, really helpful. I look forward to talking with you guys next week.