The Digital Transformation of Water

Autodesk’s Bob Bray and Colby Manwaring discuss digital twins for the water infrastructure industry.

This video was sponsored by Autodesk.

Water presents a contradiction, says Colby Manwaring, CEO of Innovyze,  now VP of Innovyze, AEC Design, Autodesk. Water is both priceless yet a fundamental human right. The challenges of delivering clean drinking water to a growing global population are daunting, and digitalization may be the only solution. 

On this episode of Designing the Future, Manwaring is joined by Bob Bray, Senior Director and General Manager of Autodesk Tandem, to discuss digital twins for the water infrastructure industry. The pair discuss the challenges of the industry, the recent acquisition of Innovyze by Autodesk, how AI is used in digital twins, the Innovyze Info360 and Autodesk Tandem digital twin platforms, the importance of cybersecurity, and how engineers should change their approach to infrastructure projects. 

Learn more about the digital transformation of water at

The following transcript has been edited for clarity.

Michael Alba: Hey everyone. And welcome to Designing the Future. On this episode, we’re going to be talking about water, both the type of water that’s made of H’s and O’s and the type made of ones and zeros. Today, we’re diving into digital twins for the water industry, and we’ve got two great guests to lead us on this dive. We’re joined by Bob Bray, senior director and general manager of Autodesk tandem; Autodesk’s new digital twin platform. And we’re also joined by Colby Manwaring CEO of Innovyze and now VP of Innovyze within the AEC design group of Autodesk, which acquired Innovyze earlier this year. Bob, Colby thanks so much for joining us on the show.

Bob Bray: Thanks for having us today. It’s great to be here.

Colby Manwaring: It’s a pleasure to join. Thank you.

Michael Alba: I think the best place to start our discussion today is maybe just to get some context on this industry. So Colby, I’m going to start off with you. Would you give us a little bit of background on the water infrastructure industry and in particular, what are the main challenges facing this industry today?

Colby Manwaring: The water infrastructure industry is one of those hidden industries that every one of us rely on but few of us notice in day to day. The main problems in the water infrastructure industry are… I’ve got a list of four. First aging infrastructure; so infrastructure is failing at an accelerating rate here in the United States. The American society of civil engineers has graded our infrastructure at a C minus grade for clean water delivery. Across the United States, a water delivery main pipe breaks every two minutes and every single day we lose about six billion gallons of water to cracks, to leaks, to broken pipes. That’s enough water for 9,000 swimming pools. So aging infrastructure is a big problem.

The second problem is an aging workforce. So the experts, the people that have been managing our water infrastructure are nearing retirement. In the next five to seven years, there is roughly 50% of that workforce that will retire taking with them a mass of tribal knowledge and experience and creating an information gap for that new generation.

Third, population growth and capital funding. This population continues to outpace the capacity of water systems and despite our best efforts to anticipate and build for the future we’re behind already. It is estimated by 2030, we will be 1.9 trillion dollars behind on infrastructure development globally.

And finally, regulatory requirements continue to increase and rightly so. The quality of our water and the safety of using drinking water, the safety of wastewater disposal or reuse is paramount. But again, if I referred to the United States, since 1980, there have been about 60 new regulations that water utilities have to comply with. That’s more than a new one every year. So it’s tough. These are challenging times to provide clean drinking water and safe sanitation and protection from floods.

Michael Alba: So obviously these are problems in need of a solution and Autodesk must feel the same way Bob, because your company just spent a billion dollars on Innovyze. So you’re clearly looking toward the future here as well. Could you give us a little bit of context on that acquisition and explain why Autodesk invested that much money and what you hope to accomplish together?

Bob Bray: Yeah, I think it’s pretty simple when you look at it and the importance of water in the world is. Everyone needs clean drinking water. Everybody needs to be protected from floods. Everyone needs sanitation systems. So the need in the world is very high, it was a gap in Autodesk’s bigger portfolio around supporting infrastructure. And so the combination of Autodesk and Innovyze seemed like a perfect match to help us build out our overall portfolio around infrastructure and also support those demands that the world is facing around water and the everyday challenges that Colby just mentioned.

Michael Alba: And Colby, maybe you could add on to your perspective on that acquisition. How did you feel when Autodesk came with that offer?

Colby Manwaring: So first off it’s fantastic to have the recognition of the value of water, albeit, through a private company acquisitions, but it’s really indicative of a larger trend around the world to recognize that, hey, this resource that we all rely on is critical and deserves more attention and deserves more investment.

So as Bob said, what is great about being part of the Autodesk company is that we can now expand that infrastructure management, or as we would say, empowering infrastructure experts across the entire asset life cycle. So from design to simulate, to build, to operate and to maintain, we can provide digital solutions, software solutions across that entire life cycle that… It just allows engineers, planners, operators to do a better job from design through to operate.

So it’s really nice to have a broader portfolio available to Innovyze’s clients, these water experts that require more sophisticated tools. It’s been a pretty obvious match for quite some time. Many, many of the Innovyze customers of course, are Autodesk customers anyway. So we saw the need in the market to provide more tightly linked, more end-to-end solutions for water infrastructure. So seemed to make good sense.

Michael Alba: And one of the key technologies, it seems to me of this new partnership is that of the digital twin. So maybe we could get to that now. I think that’s sort of the main framework for this whole partnership. Digital twin means a lot of different things in different industries. So I think we have an opportunity to really nail it down in this particular one industry. So Colby, could you tell us what a digital twin is in the context of the water infrastructure industry?

Colby Manwaring: So, I might start by stating that a digital twin for water infrastructure is not fundamentally different than for almost any kind of infrastructure. There are commonalities across the civil engineering industry. And really what a digital twin is to Innovyze and what we share with an Autodesk point of view is that a digital twin is a virtual representation, virtual data, a representation of… For water, pipes and pumps and tanks and all of the physical characteristics of those assets and a digital twin then allows us to do things like visualize the real world within the cyber world to inspect it and to characterize those systems physically. Perhaps more importantly, a digital twin includes a model or a simulation that allows us to interrogate the behavior of a water system. So again, not unlike other kinds of infrastructure, you want to know what you’ve got, where it is, what condition your infrastructure is in. You want to visualize it and inspect it and be able to present it in a virtual way. And, you want to be able to predict and ultimately fine tune that infrastructure in the cyber world, before going out and spending millions and millions of dollars building or retrofitting infrastructure.

So, at the end of the day, a digital twin for water is built on the same principles as a digital twin for buildings. Or for highways. Or for the rest of the things that, I’m sure Bob will mention, the Tandem platform supports. And it’s that commonality in digital twin vision that really brings the Innovyze plus Autodesk vision together and provides higher value to customers.

Michael Alba: So since you’re bringing up Bob, maybe you could tell us a little bit about Tandem, it’s fairly new. And maybe you could talk about if there are any unique challenges with water infrastructure. Clearly there’s a lot of overlap, but maybe there’s some special challenges there that you’ve noticed in your time working with Innovyze.

Bob Bray: Yeah, I mean, a digital twin for us, and Tandem in particular, we are a little more focused on buildings right now, but really any type of facility in general. So, thinking about a facility, a facility is made up of spaces that people occupy. It’s made up of assets, or equipment, in that facility. And most importantly, the systems that connect all of those assets together and serve those spaces. So, you want to understand with a digital twin, both the ontology, or the graph, of that system, and that set of systems that make up that building. But you also want to understand the connectivity of all of the operational data.

Most owners today, one of the biggest challenges they have, is a tremendous amount of data silos. They have siloed systems for any kind of sensors. They have another silo for operational data, like maintenance data. There’s other silos of tenant data. You really need to bring this data together and correlate that data, and really look at your data through, what I like to call, a single pane of glass.

And that’s kind of what a digital twin is initially, is bringing this data together and looking at it through that single pane of glass, so you understand all of this data. And then once you have data and you’re collecting that data, it gives you the ability to have new insights. Because you can draw correlations between the data, maybe looking at predictive analytics to say, when might this component fail in the future? When might I want to put out an RFP to replace this particular piece of equipment?

But also looking at, symptomatic problems. Maybe the problem isn’t that piece of equipment that’s starting to drop off in performance, but a maintenance activity that happened upstream two weeks ago. So being able to correlate this data together is really the power of a digital twin, that gives an owner new opportunities to look at their, the operation of their facility, much more holistically, and with a set of data that is much more available to them.

Michael Alba: I know one of the ways that you’re aggregating that data, or making sense of it at least, is with AI, artificial intelligence. Could you comment a little bit more on just exactly how you’re using AI in this context?

Bob Bray: AI is used in many contexts here, from understanding system connectivity, looking at system performance, looking at data over time to understand patterns in that data and identify problems. As I talked about, drawing correlations between data sets, AI is immensely useful there. So there’s almost every context in which AI is incredibly useful.

Even if you’re sending out a LiDAR scan to survey a pipe run, or something like that, looking at that data, you’re using machine intelligence to really understand all of that data and where the problems might exist.

So AI is in pretty much every piece of a digital twin, and touches all the aspects of taking all of that data and really making sense out of it for the operator, or the facility technician that’s looking at that data.

Michael Alba: So, that’s a little bit about Tandem. Colby, your company recently announced a platform called Info360. Could you elaborate a bit on that and maybe how it relates to Tandem?

Colby Manwaring: Well, Info360 and Tandem are twins themselves, to be honest. Info360 was largely envisioned and developed before we joined Autodesk. And so, the goal, or the anticipated use of Info360, is very much like Tandem. Bring together all the data, all the relevant data, whether it’s physical data or operational data. Get it into one place, be able to inspect it, be able to visualize it. And then, in the case of Info360, we built in the horsepower of the simulation and design engines that Innovyze has been known for, for the past 30 years.

So, bringing in that predictive analytics side, as well as some of the design side, specifically for the water infrastructure, so that we could do, I guess, advanced analytics on that platform.

But, truthfully, these are twins. They’re built on a common AWS backbone, and we envision, we envision Tandem and Info360 really coming together to be more solitary in the future.

Bob Bray: I think Info360 is a great example of a purpose-built digital twin for a particular industry. And the Innovyze team has done a marvelous job solving some specific analytics for that industry. Tandem is more of a general purpose platform.

And so, we want to embody all of that over time into Tandem, so that it can serve all of the industries that Autodesk serves in a more generalized way. That’s the way I look at it, but they are, as Colby said, they’re twins, in many, many ways.

Colby Manwaring: If I may, I’d also like to comment on how that AI and machine learning part, as it pertains to Info360. We see that digital twins need to be dynamic. In fact, we even often use the phrase dynamic digital twin. And that includes, what that means is, it includes both a predictive and prescriptive elements in a system.

Predictive means, we need analytics or models, that can take past, or even current, real time data and forecast the future. We get some sort of prediction.

Prescriptive analytics mean that, we would like that system to give us some advice about what to do about the future. And that’s where I find machine learning and AI very exciting. Many of the processes, whether they’re predictive or prescriptive, can be characterized by science. The first principles models, kind of hardcore science.

However, many of the aspects of a water system are fundamentally uncertain. We don’t know how much it’s going to rain tomorrow. We also don’t really know how a concrete pipe and a certain soil type behaves over 180 years, because we haven’t been keeping careful records. We have some indicators of that data, but simple things like how much water will the city want to use tomorrow. They’re uncertain. With machine learning, using all this data we can aggregate, we can allow the system to narrow down a lot of that uncertainty, and then our predictive analytics are better, and our prescriptive analytics give better recommendations on what to do next because we’re narrowing the assumption or the guesswork. So AI can be enhancing inputs to a digital twin. It can be enhancing outputs to a digital twin. As Bob noted, it enhances data collection that you put into a digital twin. So it’s threaded all through INFO 360, because it’s necessary.

Bob Bray: And I’ll just add one more thing to that, Colby, if I may, which is that if you think about the future, if you know what might happen and the conditions of your current infrastructure, it can also potentially self-tune, getting to that notion of an autonomous twin that doesn’t necessarily need the human operator. Yes, you want that observation of the human and the supervision of the human, but it could actually potentially self-tune infrastructure, which would be a fantastic outcome in the future to have that ability for the digital twin to self-tune, and basically self-heal systems.

Michael Alba: Yeah. It’s a very interesting goal, but imagine there’s still quite a bit of progress left to make. So what are the main obstacles in the way of achieving that end result?

Colby Manwaring: Within the world of water infrastructure there’s some unique challenges. One, a lot of the infrastructure we talk about is invisible. And what I mean is literally you can’t see it, it’s underground or inside of walls. You can’t see that stuff. So we don’t necessarily know exactly what’s out there in the world and what condition it’s in. So there is quite a bit of data acquisition or data gathering to be done, removing data from various silos and aggregating. And that’s a fairly large undertaking and it has to be done system by system. These systems are generally unique, so invisible infrastructure and just getting the data in occupies a lot of time. The second challenge I caught is nature. These systems are exposed to nature and as such, that there’s a lot of unpredictability as we try to build prescriptive and predictive models.

And I think that the last thing I’d call out perhaps as a challenge is within the water industry, it’s a little bit of the perceived value of water. Everyone agrees, water is priceless. It’s a basic human right, but those two statements are a little bit opposing. It’s priceless. There’s not enough money to put a value on how much water is. On the other hand, it’s kind of expected that we’ll just get it for free. So the challenges in funding, data collection, uncertainty analysis, and building these digital twins is a real thing, as water utilities or infrastructure owners are looking to maximize the use of their dollars that they collect through sewer fees or through water fees. In a regulated and risk averse industry, it’s a little bit of a challenge to match the funding to the most urgent problem versus the best long-term solution, right?

It’s always fix what’s broken right now, versus create a system that will help you avoid an emergency in the future. The good news is there is rapid digital transformation finally taking hold in water industry. About 60% of utilities globally have said, we expect to digitize by 2025. That’s great. Some of them, when they say digitize, they’re literally saying we don’t have anything that’s digital. It’s some things on paper and we’re going to take that first step. But as some of those challenges are overcome, there’s a great appetite for doing two things, optimizing existing assets and increasing the use of data analytics and simulation to better design the new assets that we need, and to actually not just design them, but forecast the lifetime cost of such an asset and optimize that so that we don’t get into this vicious cycle of, we need to spend all our money on building new things, but the new things we’ve built are being ignored and failing.

Michael Alba: Bob, are there any broader technological challenges that you see from Autodesk side?

Bob Bray: I think it’s more honestly not a technological challenge as it is an industry wide and a cultural challenge. As Colby said, many owners, I don’t care if it’s a building owner or an infrastructure owner, don’t know what assets they have and where they are, and what they are, or the condition that they’re in. There is a lot of assets in the world that get maintained by the seat of the pants, if you will, by the facilities team. And so one challenge is of course collecting data about the world and the built environment in a cost effective and an easy way. And that is non-trivial today. That is definitely a technology challenge. Sure, there’s lighter scanners and there’s scans and all kinds of things, but guess what? Scans can’t see what’s inside a wall or inside a ceiling and you need to understand all of the equipment.

So that kind of data capture is definitely challenging. But one of the things that we are doing at Autodesk around new infrastructure is with Tandem, of course, building a process to make that digital twin kind of a repeatable natural output of a project life cycle, so that any new facility starts with that data understanding and all of that asset record and that system record, that ontology of that, that building, or that system, or that infrastructure, but for existing facilities it is definitely a steep mountain to climb in terms of data capture. And the more we can do technologically to automate data capture, it becomes more cost effective and easier for owners to invest in that.

Michael Alba: So one of the big questions that people have when it comes to cloud, when it comes to digital twins, is the question of security. And when we’re talking about water infrastructure, that question is especially critical. So Colby, how are you ensuring the security of your platform and of the end result, the infrastructure?

Colby Manwaring: Yeah, it’s a great question. And look, this is a full-time job for multiple security experts. Cybersecurity for a digital twin, any digital twin, but particularly one that pertains to public infrastructure and digital transformation, can’t be an afterthought. It can’t be a home cooked. It can’t rely entirely on human monitoring or on physical safeguards. We partner with AWS, Amazon Web Services, to ensure maximum attention to security and attention to monitoring so that every millisecond, the system is self-monitoring for unusual behavior or intrusion. Automated monitoring and threat detection is critical to allow people to trust that their data is not only secure, but it hasn’t been tampered with in any way, the outputs they’re getting are good, and it hasn’t been breached or compromised to other parties. So it’s funny that the paradigm in the past has been, “Let’s isolate our data to make it secure, put it on a server, put it in that room, don’t hook the internet up to that computer and that data is secure.” Well, it’s not. I have a USB stick, I can get it. And when it is too secure, it’s also useless. So it’s a question that’s on everyone’s mind. But again, I’d go back to say, look, there are literally thousands of people working to ensure maximum data security on the cloud backbone that we use. And as such, I think we say with high confidence that we can provide that right level of security for our water infrastructure.

Bob Bray: Yeah, and I think generally at Autodesk, we have been building software in the cloud for years now. We’ve actually built an entire security team, that focuses on working with our product teams to ensure best practices, to ensure we’re constantly testing and measuring, implementing new security schemes to make sure our software can be fully trusted and is fully secure. It is a full-time job for many, many people, and it is not an easy job, there are new cybersecurity threats every day. And having those teams of people that are constantly aware of new threats, building new safeguards, and working with our product teams to ensure product security, is the best path to that. And I think for owners, AEC firms, out there listening, it is a lot of effort to ensure these systems are secure, and definitely encourage you to lean into your partners, understand what we’re doing, and the best practices we provide, and I think you’ll find that our software is best in class when it comes to security.

Michael Alba: That leads me into the final question I wanted to ask you both today. With the availability of digital twin software, like Tandem, like Info360, does that change how engineers and designers should approach infrastructure projects, right from the beginning?

Bob Bray: Absolutely. So we have talked about in the industry for years, and with BIM in particular is where we started this start with the end in mind. You need to understand the data requirements that the owner has, the desired outcomes that an owner has, so that you’re capturing that information through the design and construction process. And basically, that digital twin is that natural deliverable at the end of the project life cycle. And as an AEC firm, there’s distinct opportunity to provide more value to your clients, and create a longer, more meaningful relationship with that client through data. And I think having those conversations, first and foremost, about data requirements, understanding those data requirements, how those can be delivered to an owner in a meaningful way, and then actual, they putting the processes in place too, as I like to say, start digital, stay digital, and deliver digital. So that’s a big thing on the engineering side.

Michael Alba: Colby, do you see the same?

Colby Manwaring: If I could chime in on that, I think it is important that it changes the mindset from, I guess, project start, project finish, and onto the next project, to creating a persistent data model that can be used beyond the end of the project. Because then the next person picks up and does their project, and the next person picks up and does their project. If it’s built on a common persistent data model, then the mindset is not just do it and deliver it and forget about it, it’s, “Oh, I need to make sure that this digital twin is in a condition and in a state that the next expert can pick it up and use it.” And we get then into a circle of design, build, operate, maintain, and then design, then retrofit, build, operate, maintain. And around and around it goes, so it becomes circular rather than a linear project life cycle.

Bob Bray: Yeah, I’ll chime in one more thing on that, which is, we talk a lot about data driven design and generative design, and all of those algorithms can be much smarter if we have a better understanding of how assets perform in the built world. So that circle that Colby just mentioned is incredibly important. Take that operational intelligence about the way that asset works, and feed that back into the design, and plan, design, build life cycle so that we can plan better assets and better buildings, we can design them better for the conditions they’re going to be in, and we can construct them better, really achieving a better future for everyone.

Michael Alba: And it’s an interesting and necessary new paradigm. Colby, Bob, thanks so much for coming on the show today. It was a pleasure speaking with you both.

Bob Bray: Thanks for having us.

Colby Manwaring: Thank you.

Michael Alba: And thanks to you for tuning in, we’ll see you on the next one.

Written by

Michael Alba

Michael is a senior editor at He covers computer hardware, design software, electronics, and more. Michael holds a degree in Engineering Physics from the University of Alberta.