What the Heck is a Digital Twin? (feat. Issa Ramaji)

Are you training? Feta. Randy Collins Construction Brothers

Listen now or check us out on one of these platforms!

Apple Podcasts


Google Podcasts



This week we had on Issa Ramaji. We dive into digital twins and how they are more than a snapshot of what the building looks like. We talk about the functionality of digital twins and how they bring value to the table. We look at the move towards automation of the monitoring and operation of a building and how digital twins enable us to create historical and real time data about the functions within the building.

Text us and let us know what you thought of this episode! 478-221-7009


Issa’s LinkedIn

Please consider subscribing to our podcast!


Like us on LinkedIn!      

Like us on Facebook!    

Follow us on Instagram!

Eddie's LinkedIn      

Tyler's LinkedIn      

(Our day job)   


Tyler: (06:42)

Better building life cycle, better health of the people within the building. That was mind boggling to me is being able to read different rooms and basically the air quality inside of them. In schools where our kids are at. It's being able to make sure that we're giving them a clean environment to be in. I just, I think that's something that is going to continue to blossom over time as the internet of things, the IOT, starts to develop and become more prevalent as the years go on. Well guys, this is a fun conversation and we're going to get right into it. So here is our conversation with Issa Ramaji.


Alright well, Issa thank you so much for joining us today. Can you tell us who you are and what you do?


Hi Tyler, it's a pleasure to be here today. I'm Issa Ramaji, I'm the CEO and co-founder of dataArrows. I'm also an assistant professor at Roger Williams University. I've been in the building industry for about 14 years and I've been all over the place. I've done design, I've done construction, I've been in academia, and my most recent work has been starting a company from the ground up.


Well, we wanted to talk to you a little bit about what you're getting into with that company and that is digital twin. So for starters, what's a digital twin.


Yeah. That's a great question. Digital twin is well, there are many different definitions of digital twin in the literature and there is no one agreed upon definition for digital twin. If I want to put it simply, I would say it's a digital representation of a physical asset. And so with various kinds of purposes, you might have a digital twin for monitoring something. You might have a digital twin for controlling something or any other purposes.

So it's a digital representation of a physical asset. And this is not a new concept. We have been using digital twins for decades. Like a very simple example is your car dashboard. Have you looked and paid attention to what information you get from your car dashboard? It gives you real time information, shows your speed, your mileage of your car. It shows you some predictions like how much further you can drive with your car with the gas you have in your car. So that's a prediction of what's happening in the future. It gives you alerts on your engine, a lot goes on, that's an alert. So, that's a digital twin. It's a digital twin of your car sitting in front of you. So it's not a new concept. And so, but a lot of innovations have been incorporated into digital twin. And so I think that's part of not having a unique definition for what digital twin is because it's evolving, it's evolving technology. So we are just exploring the potential of digital twin. So I think that's part of the reason there is no agreed upon definition for that.


I love that comparison to a car dashboard. I think that really kind of helped just kind of explain things a little bit further for me in the digital twin realm. So I constantly think about BIM and what we do on a day-to-day basis as being a digital twin, but not necessarily to that level of its operational status, which was really the thing that you and I kind of explored whenever we first got on our first phone call was, here's the data that you can pull from these models and learn about how's your air quality? How has this, how has that? So, let's dive into the benefits of digital twins. What are some of the benefits? What are people seeing out of these that are justifying the cost of creating one?


Yeah, well, we should talk about benefits in the context of use cases. Like we had BIM. BIM has been very successful with delivering lots of benefits for design and construction. I think the kind of new benefit that digital twin can offer to our industry is mostly related to the operation side and then learning from the operation and previous projects for doing new projects. I think that's just better operation, more efficient, more smart operation of facilities. And they're learning from that for the new assets we've built. I think that's where the main benefit of digital twin happens in my opinion. And again, like if you can use it, we have had successful examples of implementing digital twin for control of building systems, like your HVAC system. We have had examples of monitoring systems. And a more kind of recent application is the integration of AI with digital twin. We are also getting to the autonomous operation of building systems. So I think a lot of new unexplored areas exists in the operation stage of the building life cycle that digital twin can contribute to


I've thought of digital twin as just maybe being a capture/a picture of a building. And I suppose in a sense it is, but what you're describing sounds a lot more like the retention of information in order to use that for future benefit. Why don't you tell us more about how that information is the main lever in this thing that it's not just maybe taking pictures of your site?

Issa: (12:48)

Yeah, absolutely. So I've been in design companies, I've done consulting for several years, and we do a lot of advanced computational analysis analyzing building before we build. Like, we do very complex energy analysis to predict what would be the energy demand of a building or energy performance of a building. But do we ever look into what happens after the building is constructed? All those assumptions and all engineering analysis incorporate a lot of assumptions and simplifications, so we never had a kind of mean to do that. We build the buildings, owner's happy, everybody's happy, we get to start a new project.

But digital twin enables us to capture information monitoring for real time at the same time store historical data. And then having stored data, we can compare what we predicted, what is the actual kind of performance of the building. And from there we can go and tailor our design assumptions and learn from them and take it from there. So that's why I was saying digital twin is not only for operation, but also it can add a lot to the design stage of a building life cycle. Because again, this is a personal experience but when I did my PhD at Penn state in the architectural engineering program, and an AE program at Penn is one of the top three architecture engineering programs in the world, we were learning all those advanced technologies. But in my office when I was PhD student there, we used to open windows to control the temperature in our room. So we were learning a lot of advanced solutions technologies but we were operating buildings like that. We're literally contributing to global warming by opening windows and letting warm air go out. So, I think that's where it was missing. So we need to take knowledge and all these technologies and solutions we have developed to analyze the building performance, to the operational side. And I think digital twin could be a good mean for that.


I think it's really interesting, the historical data that you're talking about. And this is one of the interesting conundrums that I see happening over the next 20 years we'll say. Is that everybody, say your company, will have another company over here and another company over here. Everybody's collecting this data, this historical data about buildings. Eventually, all of this stuff is going to have to come together so that we can make decisions off of it. And so the sharing aspect of things really has to start changing, I think, long-term. So that's one of the things that I think is going to be a hurdle over the next 10, 15, 20 years, as far as all of this data getting collected and sharing it out. I don't know if you've seen anybody pushing the stuff out there. So it's more open source, or if this stuff is still kind of locked down to you guys, what are you guys kind of doing in that regard to share this information and help?


It's a great point. So you're talking about learning from your data, not just to benefit your company, but to kind of benefit the whole business as one unit. And I think that the need for open communication and open standards has been kind of understood by the industry. And we have had this like open technologies effort for developing open technologies for BIM. I think this has been continuing to digital twin and the need for open technologies and open platforms is even much more profound for digital twin because of the variety of technologies that are involved in digital twin. Just think about it, like BIM with all its challenges, it's just one component of digit twin and this is true if you are not just working with as put information, you have to work with sensor data. You have now IOT involved in this, and then you have artificial intelligence involved in this. Then you have your city information involved in this if you want to take digit twin to smart cities. So then you find yourself in a much larger kind of ground that a lot more technologies are involved, even various disciplines. And then now the need for integration of information becomes more profound. And then I think that's a need definitely a need. And I think that is being understood already by the industry. Now, it's just a matter of time that we develop those standards for the industry as the technology itself is evolving.


It's definitely going to be something that affects our designs from an architecture standpoint moving forward. Right. So if we have that historical data to back up these design decisions that is going to be, I hate to use the word game changer, but I'm sorry. I'm sorry. But no, it really will influence the way we design structures going forward is having that information at our fingertips. Now on the sensor side of things, let's kind of get back into how you're getting this data out. What kind of sensors are you implementing? What does this look like for implementation in the field?

Issa: (18:51)

Yeah, that's also, you're pointing out one of the major challenges with digital twin integration with sensor data where I mean, let's put it that way, integration of your live data with your static data, which is as built information, I think that's one of the major challenges. And I mean, it depends on the use case. If you want to use it for structural monitoring, you have structural sensors. If you want to use it for health assessment, you use indoor air quality sensors. And one of the issues with sensor data is there is no standard. We have communication standards for sensors like some protocols that communicate data, but architecture of the data itself, there's no standard for that. And I think that's what makes the integration of various sensor data in one kind of integrated platform challenging. Does that answer the question?


Yeah. Yeah, it does. And, so from a...I guess if I'm going out and I'm trying to add these sensors to this building, where should I look, where should I buy these sensors? I'm sure there are multiple companies out there too, but if I want to put a sensor for my structural package, for my steel, what are some of the areas that I can look or where can I look to find some of those different assets to place?


We shouldn't separate the kind of sensors that can be used with digital twin from the sensors that have already been in the market for years. Sensor is sensors. I think the missing part, the missing piece, of the puzzle here is integration of data that those sensors are providing with your digital twin platform. So what we have done, we have been active in indoor air quality and indoor health assessment. So we have been working with IQ sensors, indoor air quality sensors, a lot. The way we have addressed it is creating that interface. We create the interface that can read data using various communication protocols from Bluetooth to LAN or even backnet. And so, I think the missing piece is that interface from various sensor technology with IOT platforms, which are the base of digital twin technology. So I think if we have that interface and integration of sensor data with as-built information it's going to be a lot easier.


I have something that I'm thinking of while you're talking too and it's that integration aspect is so huge. And I don't want that to be understated as well but so, on my phone. So I love smart home devices, so nobody go out there and hack me, right, please. But I love smart home devices, things like Phillips hue. I like my wifi router easily accessible. I like my nest thermostat. I like my simply safe alarm system to be all kind of housed on my phone and all that. But the problem that I've kind of unearthed and before this call even, I was sitting and fiddling with the Phillips hue app, trying to make sure that my lights were connecting right. But the thing is, I've got an app for every single one of those systems and it becomes a lot to manage when you start scaling it up. Now, mind you, this is just in my home. So imagine that you're working in a structure, that's a hundred thousand square feet, this gets maddening. So it's great to have that stuff come back and tie together. I just didn't want to understate that point, because that is so critical when you've got sensors and things that are reading that data.

Issa: (22:56)

Yeah I think, going back to your point, having open standards for communication of those data is crucial. And then that's like what I've been trying to achieve in digital twin consortium. I'm leading the data interoperability group and digital twin consortium. And that's all we talk about. When I want to see, "okay, what technologies are needed for digital twin and how can we come up with one simple interface that if all companies can implement," we can have a seamless and easy integration. It's easy to say. It's very hard to implement. I mean just going on the BIM side of digital twin, I mean, people have been working on it for more than two decades now. Just creating some open technologies that no matter what BIM tool you use, you can pull all your data in one platform.


And a lot of good progress is made, but we're still not there. So a lot of things are missing. So I think it takes time. It takes, unfortunately, it takes time to have those standards in place, but the good thing is people understand that need. So, I mean, just going back to this digital twin consortium, so the whole concept of digital twin became trending maybe two years ago, but people are jumping on the need for interoperability and open standards because they have learned from previous technologies that this is the need. We better think about it as the technology is evolving.


So I want to be kind of the simpleton in the room because I play that part well. Going back to the beginning of the building life cycle, we start in maybe development and then get into design. And the design gives way to the construction processes and those construction processes give way to building life cycle processes. And each one of these bus stops along the way has maybe a different facet or tenant of digital twin associated with it. So we are equally describing digital twin as the information that I can associate to a structure from BIM as the information I can later associate to that structure via data that I've gathered from a sensor. And so this is the gathering of information. It's the things that I know about a structure that I can therefore leverage later and use to better build the next one, correct? Am I understanding that appropriately? That digital twin has all of these things wrapped into one, not just maybe reality capture during the process of construction?


Yeah, absolutely. Absolutely. It's all about being able to generate the information throughout the life cycle, the whole life cycle of a building, keep a record of that, create historic data, and learn from it. You can learn from it for your design. You can learn from it for autonomous operation of your system. So being able to capture data in real time and learn from it, I think that's where the novelty of digital twin use cases would look like. Just being able to reuse information for various purposes. I mean, not all this information should be in one platform, so don't take me wrong. I'm not talking about having a black box when we put information from various sources in the box and then have everything intergrated. Now that's been practiced with BIM, never happened. It was unsuccessful. I'm just talking about capturing the information and data you want and use it for a specific use case. While having these use cases somehow not integrated, but what we need is several use cases for digital twin that have an interface with each other. That's what we want, not one black box, not a data lake. We don't want a data lake because it's very hard to achieve. And, I mean, I think we have learned from BIM that it's impossible to achieve. So what we want is several use cases for BIM, and for these use cases to have an interface with each other so they can communicate the information that they might want from each other easily.


I want to kind of veer back into some of the practical application of this, right. So do you have any stories that you could share with us from people that have implemented a digital twin and they've benefited from that in some way or another, do you have any stories around that?


Yeah, absolutely. So the digital twin enabled us to do things that were not possible before. I'm just gonna give you one of the use cases we have recently kind of been implementing since COVID happened. And that was the need for assessment of indoor air health. Traditionally, without digital twin, the way this assessment is done is you're testing your building every few months to see what's the condition, what's the performance of your building system. Or having one CO2 sensor on a wall, just seeing what's the CO2 level of the room. And none of those things can kind of guide you to the right direction, because a lot of other factors are involved in assessment of indoor health. Let me give you an example. Being in a room, a very small room, three square feet, very small. One person in a room, if the CO2 level is 1500 in that room, you wouldn't be as concerned if you have one person in a large room, but the CO2 level is 1500. So it means the ratio within the occupants space to volume of the space matters when you want to talk about ventilation of your space. So what we have done is we wanted to kind of look at how different health agencies talk about ventilation and healthier indoor conditions. And we realize they are all speaking in the language of air change per hour (ACH). How many times you change the air of indoor air/indoor spaces. So if you want to calculate that and there are a lot of factors involved, you might have your windows open, you might have your HVAC system running. You might have a variable number of people in the room.

So a lot of factors involved. Can you do it with a simple sensor? No, because all you can measure with a sensor is measuring CO2 levels of the room. You cannot address the volumes of the spaces. Can you do it with BIM? No, because you want sensor data. So what we did in that application was integrating building geometry with sensor data and through algorithms we were able to calculate ACH. So now if you can enable facility managers to measure ACH, then they can easily compare the ACH of the room with the guidelines like CDC or Ashrehab. And then judge if they are conditioning their building correctly or not. And that has been a challenge for a lot of facility managers when talking with customers. People have been spending hundreds of thousands of dollars on renovating and upgrading HVAC system. But I didn't know if all those effort has been enough or not. Have they been doing it more than what they need? Have we been overdoing it or not? But digital twin can enable a facility manager to have a realistic look into their building condition, and then operate it based upon that information. So that's a good example I would say for the benefit that digital twin can deliver that would have been impossible to deliver with sensors alone or BIM alone.


What is the greatest thing you see coming of this and say 20 years?


I think the autonomous operation of building systems, I think that's where...20 years is not a long time line. So, yeah like BIM is 20 years. I mean, can you believe it? It still feels new, but it's been 20 years. So I would say autonomous operation of building systems, not building systems, I would say autonomous operation of building systems that are connected to a smart city system. I think that's where the digital twin and these smart solutions are heading.


And then the growth of the IOT side of things, the internet of things, the growth of that, it should only serve as a catalyst to help digital twin, I would think?


Yeah, I totally agree with you. I mean, IOT has evolved to a very mature level right now. I mean, IOT is very mature technology. It's not new anymore. So I think like we have incremental development and improvement in IOT technologies that are enabling more applications. Like with 5g now you can have internet access at the city level, and then that will enable even more new use cases for digital twin.


All right. Well, I think we've come to a great spot to ask you our mega phone questions. So buckle up. All right. So if we gave you a megaphone that the whole industry could hear in around 60 seconds, what would you want to say?

Issa: (33:41)

I would say if you want to use and leverage a technology in your project, just analyze the benefits that the technology can bring into your project first and then implement it. So all of us in the industry are so excited about the new technologies that are being available to us. The technologies that we couldn't have even imagined five years ago are becoming accessible to the industry right now, but using technology doesn't necessarily add value to a project. Like going back to just give you an example, BIM, we don't see much adoption of BIM in residential construction because people have tried it and realized financially it doesn't make sense at the current stage to implement BIM in residential. So if we can handle projects without a specific technology and that technology can not add much value to the project and it will just cost more. I think there's no need for using that. So I would say my answer would be first to do an analysis on the value that the technology can add to your project and then adopt it. I mean, don't take me wrong, there's always value in trying new technologies because we can learn from it and all new technologies need some early adopters, just giving new technologies a try in spite of all the flaws that the new technology might have. That's very important, but I think it would be good to try new technologies on the projects that they can potentially add value to. Not just trying it for the sake of trying a technology.