Episode Transcript
[00:00:03] Speaker A: Welcome to AINA Insights, where prominent leaders and influencers shaping the industrial and industrial technology sector discuss topics that are critical for executives, boards and investors. INA Insights is brought to you by INA AI, a firm focused on working with industrial companies to make them unrivaled segment of ONE leaders. To learn more about INA AI, please visit our website at www.aina.AI.
[00:00:40] Speaker B: Welcome to another episode of the Titanium Economy podcast hosted by aina.
Today we are joined by Riaz Rehan, Chief digital Officer of Trane Technologies.
Trane technologies is a $21 billion global climate innovator, maker of the Trane and thermoking brands, serving customers across commercial buildings, homes, transportation and industrial refrigeration in over 100 countries.
The company has staked its future on net zero across its supply chain by 2050, with hard interim targets along the way.
Riaz is an unconventional path. A mechanical engineer by training who spent his career building digital and software businesses, leading IoT at Cisco, digital engineering at Infor, and serving as CEO of a $1 billion SAP subsidiary before bringing that fluency back to the world of machines and physical systems.
Riaz holds engineering and MBA degrees and is a graduate of the Stanford University Management Program. Riaz, great to have you with us.
[00:01:47] Speaker C: Thank you, Mohit. A real pleasure to be here.
[00:01:50] Speaker B: All right, Riaz, so given this is the Titanium Economy podcast and your prior experience building these businesses at Cisco, at SAP, at Infor, how do you feel or understand Titanium Economy and what attracted you to this sector?
[00:02:06] Speaker C: You know, it's a really good question. And for me, industrial manufacturing in the industrial sector has always been the heart the US economy.
And quite frankly, any industrialized nation and everything that happens around it is of great interest to me. Having worked in IoT, having worked in companies like SAP, which primarily serve manufacturers, it was always an area of interest to me. And as you pointed out, I'm a mechanical engineer by training. I think always a mechanical engineer at heart.
So it was a natural extension of my career. It went up in the second.
[00:02:41] Speaker B: Got it, got it. That's interesting. And then coming to train technologies itself. Right. It's a pure play climate innovator and it has bold sustainability commitments. The Gigaton Challenge, for example, net zero. By 2050, about 15% of global greenhouse gas emissions come from heating and cooling buildings. Right. So how does digital turn that commitment into operational reality rather than just an ambitious target?
[00:03:09] Speaker C: On paper, yeah, we take our targets very seriously. And the Gigadon Challenge is something that we are actually on track for. In some cases, we are slightly ahead of track.
[00:03:19] Speaker B: All right.
[00:03:19] Speaker C: And we do this in multiple ways. But I'll talk about specifically how digital enables this.
I'll give you a couple of simple examples.
One of the big parts of our business is service. About a third of our business in the commercial H VAC sector is services.
And if you think about services, one of the biggest kinds of services we do is maintenance services.
Today, we use digital technology to stay connected to some of the largest commercial machines. And this avoids us from making truck rolls. And if you multiply that across the length and breadth of our company, that's literally hundreds of thousands of truck rolls that have been eliminated because of digital technology. So a simple but very easy to understand way to see how digital can drive sustainability.
The other big one, of course, is artificial intelligence. We recently acquired a company called Brainbox, which think of it as AutoPilot for commercial H VAC systems. Brainbox uses structured and unstructured data to optimize how these systems operate. And without sacrificing comfort, you can reduce your energy consumption by 25 to 30%. So there's things we do in how we run our business and there's things we do to make our customers, businesses more energy efficient and more sustainable. And of course, there's things we do to make our. Make our company more circular. So all of those combined, I think, is a. Is a great way to understand how digital impacts sustainability.
[00:04:47] Speaker B: Yeah. So I think you bring up two very interesting points. Right. You mentioned that one third of your business is digitized services. Right. So how do you see in the future this service delivery being augmented by the digital delivery model that you're updating? Does it change fundamentally what train technologies does and is?
[00:05:08] Speaker C: I think digital technologies are having a fundamental impact on how services are delivered.
I'll give you a very simple example.
For several decades, the manner in which large commercial H VAC and even residential H vac systems were serviced was based on a truck roll. It was based on a technician, a human, showing up, typically for a commercial system, four times a year. So each time the season changed, or for a residential system, two times a year, and performing activities after doing an inspection of the system. And some of these activities were fairly simple, changing a filter as an example. But others were more complex.
And the fundamental premise was that you would have a technician making the truck roll and showing up to the system.
Now, with digital technology, with IoT type technology, we can reverse that. We can say, first of all, why wait for the seasons to change?
Why not have sensors on the equipment that are connected to A modem and have that modem connected through mobile technology 5G or LTE.
Why not monitor that equipment 24 by 7?
Why not be in a position where you have exception based monitoring? So when a certain metric or a certain quantity or a certain parameter is out of range, before it gets to red, it gets to yellow. And when it gets to yellow, we can proactively let the customer know this is happening. So we know the issue has happened and have taken action even before the customer knows. And then you roll a truck when you need to roll a truck. And as opposed to it's four times a year, the season has changed to roll a truck. So it's a fundamental change in how we deliver service without compromising on quality. As a matter of fact, one could argue that the quality of service delivery has gone up, the cost to deliver the service has gone down, and customer satisfaction has actually gone up as a result of that. So it's a triple win. It's good for the environment, it's good for the customer, it's good for business.
[00:07:09] Speaker B: Yeah, that's amazing.
I think the second thing that you mentioned, which was very fascinating for me, was the acquisition of Brainbox and the whole AI play that is there. Right. When I talk to CDO CIOs within industrial companies, I think they're still in that phase of what should we do with AI? Where should my investment be? Should I grow a team internally? And there's a bunch of communication there, but everybody seems to be at or near the starting line. And Trane had the foresight of investing in Brainbox last year and now it is scaling, it is being commercialized. So walk me through the thinking, like, what led you to that decision much earlier, much farther ahead than the rest of the industry, and how did that culminate into the product that we see today?
[00:07:52] Speaker C: Yeah, so we've been strong believers in the potential of artificial intelligence to solve real world problems.
Not problems that are imagined or problems that are fake, but real world problems. And if you look at the H Vac business, especially on the commercial side, one of the biggest challenges is when you commission a building, which means you've set up a building, you've installed a new H Vac system at the point of commissioning, it has a certain efficacy and efficiency which is near perfect. And it kind of deteriorates slowly after that. So in a nutshell, Brainbox AI allows us to reverse aging and to keep that building in a state of constant commissioning. So it's always as born.
Now, this is easy to say, but Difficult to do. And one of the big things we do is we work with the controls of these very large, complex commercial H VAC systems. Because we sell systems, not just machines.
And those systems include machines, they include controls, and they include a whole lot of services. But essentially, you have these controls that are managing massive machines. Some of these machines individually are the size of a school bus. And most buildings, most large buildings, will have many of them in series and tandem. Operating the H VAC for the building, there's a lot of control settings that. That can be changed to match the requirement of the building in that point in time.
Now, typically, what happens is when you set up a building, when you commission a building, you're making these decisions based on certain assumptions, and those assumptions are accurate at the point of making them. But then, like anything else, things change over time, and no one goes back and changes those settings constantly to match those changing variables. Yeah, that's where artificial intelligence comes in. So the reason we like Brainbox is because Brainbox was using machine learning and using quantitative predictive analytics to make really good decisions on how to change those control settings on an ongoing basis.
I often use the analogy of comparing Brainbox to an autopilot in a commercial airliner.
[00:10:12] Speaker B: Right.
[00:10:12] Speaker C: The autopilot is making tiny adjustments to the flight path of the aircraft.
And this might surprise some of the listeners, but most planes during flight are pointing in the right direction less than 5% of the time. So 95% of the time, they are pointing slightly in the wrong direction. And it's the autopilot that makes those tiny corrections, often a fraction of a degree, to get the plane back on track. When you flew in from Houston, and had it not been for Autopilot, you would have ended up in Boston, not in Charlotte. And.
And that is what Autopilot does. It's those tiny changes. Brainbox does the same thing, but for H Vac it's tiny changes.
And of course, we can discuss in more detail how it pulls the data that allow those changes to happen. But that's the basic idea. And we liked it so much. We knew this would be of big value to our customers. We knew that we could scale it using our massive install base and using our salesforce and using our technician force, which is why we made the acquisition.
[00:11:12] Speaker B: I'm very interested in the start of this whole process because I think the end product, I definitely agree. Right. This is something that will revolutionize, change how these systems work. But when you start this process, do you have a thesis in mind that I want something that will allow me to make these fine tuning adjustments to match PAC systems. Or do you first come across a company like Brainbox and think about a use case that, hey, this is where I could apply this and this could be a game changer for me. So where does that thinking process start?
[00:11:40] Speaker C: There's always been an aspiration to have this kind of capability, but ideas are out there. There's lots of ideas. Execution is what counts. And as we look, we looked at a number of companies and we looked at them, we partnered with them, we had discussions with many companies and we thought that Brainbox really had a solution that worked. Got it. So the first thing we did was we signed MoU with them and we started partnering. We picked a couple of our key customers, and there was one customer in particular that had multiple sites across the United States and Canada.
We installed Brainbox. We ran it for several months, we saw the results, we spoke to the customer directly, we got great feedback, and that became the basis of converting that MOU into something more serious. And eventually that led to the acquisition. So we certainly had the aspiration, but we also, we also used a substantial amount of time to test the hypothesis, to make sure that we were acquiring something tangible and real.
[00:12:40] Speaker B: And I think you bring up a really good point because I think a lot of times what happens, especially when people are developing AI use cases, you go into this pilot mode that I need perfect data and I need perfect solution and you create a pilot, but it goes nowhere. So you actually did a real world application and saw if it actually would work or not and scale it from there. There, that's excellent. And how did that partnership that then evolved?
[00:13:03] Speaker C: So initially we worked with a few customers together. They were trained customers and Brainbox customers.
We learned a lot. We learned how to pull structured and unstructured data. And I think that was a big learning.
The structured side is less complicated than the unstructured. So you're looking at things like the layout of a building, you're looking at the equipment in the building, you're looking at things like occupancy levels.
And the unstructured is things like cloud cover, which changes constantly. Weather forecasts, which are very complex and notoriously difficult to predict.
Particulate matter in the building, around the building, because all of these have a role to play in the outside and inside temperature of a building.
Listeners might be curious to know that in commercial buildings, oftentimes the same building is being heated and cooled at the same time because they're so large and they have different sides facing the sun and away from the sun. So it's a very complex mini ecosystem that you have to understand.
The advantage of AI is it doesn't sleep, it doesn't get tired, doesn't take vacation, and it's 24 by 7. So we take all this data, feed it into the cloud in a very secure way, and the algorithms of Brainbox go to work. They are constantly predicting what will happen on a rolling six hour into the future and predicting what the zone by zone temperature will be needed in that building six hours into the future. And then that becomes an input into using, making the changes to the settings. And these are all minor changes. So eventually think about it, landing a plane very softly into that future by knowing what the future will be and doing that on an ongoing basis. That's what smooths out the operating curves and that's what gives the energy savings.
[00:14:53] Speaker B: Got it. That's amazing.
I want to start looking at how AI is getting applied internally in train as well. Right. So you've framed Tein Technology's AI agenda around three lenses, right? You've said AI for growth, AI for business transformation and AI for employee capability.
Walk us through that framework. Right. Where does Trane Technology stand on each today?
[00:15:17] Speaker C: So it's a very simple way of looking at AI.
AI for growth we just talked about. Brainbox is a great example, but it's not the only one. We have AI in another company we acquired called Novolo, which uses AI to make enterprise asset management seamless, easy, almost fun. We have AI embedded in trainsupply.com, which is a website that does billions of dollars of revenue for us. And AI driven search is helping drive growth. So AI for growth is a very important pillar because this is how we drive the top line of the company. And we are a growth company. So growth companies need AI to drive growth from a revenue standpoint. Let's move to number three and we'll come back to number two in a second.
But AI for employee experience I believe is where a lot of value exists at the individual productivity level. And what I mean by that is, is allowing and enabling our employees, our colleagues, to use AI in their day to day work. I think it's very important people are using it differently. Some people use it to summarize meetings, others use it to rewrite documents, others use it to do research.
Some people I know are using it to fine tune job descriptions. There's lots of little use cases that proliferate across the company. And we have seen literally tens of thousands of people in our company Using AI. But this is all about individual productivity and we think this is important. It's like email. What's the business case for email? The business case is it makes everybody more productive, but you can't actually package it up and put it in a P and L. Right. But it's very important and we call this AI for employee experience.
We encourage it, we have enabled it and we think it's really important. Now that leaves something in the middle, which is AI for business transformation.
There are certain use cases inside any company, especially a manufacturer, where AI can be used in a very targeted way to solve very specific problems. Where you need the AI vendor to have not just the algorithms and the model, but the context.
I'll give you a couple of examples. One is easy example that anybody can relate to is freight.
Optimizing freight for large multibillion dollar manufacturers that buy tens of millions of dollars of freight is a no brainer. And there are good companies out there that have the domain expertise and the context and have applied AI to solving that problem. Now when you apply AI for that, it can lead to a very direct benefit that can have a P and L impact very immediately.
[00:17:53] Speaker B: Yeah.
[00:17:54] Speaker C: Another very easy to understand example is contract intelligence.
A lot of the contracts that companies sign have rate rebates, entitlements and many other negotiated terms that are trapped in a PDF.
But if you look at their system of execution, their ERP, whether it's SAP or Oracle or PeopleSoft, all of those terms and entitlements and rebates might not be reflected in the system of execution. That leaves a huge gap. AI is very good at fixing that problem because it can use Genai to read the PDFs and then it can use Agentic AI to make those changes in the system of execution. Again, this could be multimillion dollars for most companies. A third example is inventory optimization. If you think about understanding demand and understanding demand for forecasts and using that to drive manufacturing plans and using that to drive procurement plans, there's opportunity there as well. So these are the kinds of ideas I think manufacturers can deploy AI to drive business transformation. And in each of these cases you can see there's a very direct link from using AI to solve a problem that leads to a financial metric that can be measured and then can hit the bottom line. So that's how we look at these three pillars. And for us, this is our AI strategy.
[00:19:13] Speaker B: Awesome. So I think there's a couple of interesting things which we can take away from this. So you mentioned AI for employee experience. Right. And that allowing the employees to use AI to enhance their own productivity. I think some of the things that industrials especially grapple with, how do you unleash innovation but still manage the risk, given how this data is going to get out in the world and how people are using it, et cetera. So how have you set up a framework which optimizes that for you? And within train and then I think on the business transformation side as well. Right. So there's multiple use cases. I'm pretty sure once you start brainstorming, people come up with ideas. Right.
How are you governing kind of what comes first? What is the priority? Where would we invest resources, given the resources tend to be finite with these things. Right. So how have you set that governance up and how are you thinking about it?
[00:20:08] Speaker C: So I think companies have to understand that when you talk about employee experience, everyone's experience is slightly different.
So in employee experience, you want a many to many type approach. You want everyone to try and encourage them to use the tools and let them figure out their own path. And people are very creative in figuring this out. There's people who talk to each other. Different people will use the same AI differently.
Different people will have different use cases that they will share with their colleagues. And we encourage that. And that leads to a plethora of mini use cases that have varying degrees of adoption.
And we think that should be left unfettered.
You should allow that to grow and you should allow people to use it as they see fit, but not try to count the benefits. In financial terms, that's about experience.
Right. What's again, what's the business case for email? It makes everybody's life easier. Right. Don't try to monetize the benefit of email because quite frankly, there is no alternative. Today AI is the same way. Everybody should be encouraged to use AI and that should just become the new way of working. Right. However, when it comes to business transformation, you have to take the exact opposite approach.
You have to start with the P and L. Yeah. Start with the balance sheet.
This is not about crowdsourcing ideas.
The crowdsourcing can happen on the employee experience side.
[00:21:34] Speaker B: Got it.
[00:21:34] Speaker C: But on the business transformation, it has to be a hard, financially driven.
Pick a few use cases and go hard and deep.
So fundamentally different approach on the employee experience.
Let them try whatever they want to and let them experience and learn and let it evolve organically. When it comes to business transformation, the exact opposite. Pick a few, go deep, drive the value, and start with the ones that are the biggest in terms of Value for most manufacturers. Those are things like procured goods and services, Those are things like freight.
And those are the things that one should go after inventory. Those are the big ones, right? Don't crowdsource, go after where the money is.
[00:22:14] Speaker B: Brilliant, Brilliant.
So another common question that comes when we think about buying AI, especially in industrials, right? So when I talk to CIOs, CTOs, right, one of the common pushbacks is that I'm waiting for perfect data, right? So let me transform my erp, let me set this up, and this will make sure my data is clean. And then I will start on this whole AI transformation journey.
Based on your experience, right? When you joined Train, what was the data infrastructure like? Did you have to work through it? Like walk us through kind of that transformation?
[00:22:48] Speaker C: Perfect data is a myth.
It's like the Loch Ness monster, right? Doesn't exist.
I mean, I was in technology for a quarter of a century before I came to train.
And I've had the privilege of working with, I'd say, a large percentage of the Fortune 500, and none of them has great data.
Even the great hyperscalers, many of them who sell search capability, don't have great data. So great data. Perfect data is a myth.
And I think we have to be comfortable with that. We just have to work with data as it exists. That's the bad news. The good news is that there's great tools today, many of which are AI driven, that allow you to take this imperfect data to drive really good outcomes.
The approach we have taken and the approach I think industrials should take is the Pareto principle. Focus on the 20% of the data that drives 80% of the value.
For example, there's no point knowing everything about all your products in your catalog. It's really focused on the top 20% of the products that have 80% of your revenue and accept that there's a long tail that you don't need to go after. So that's kind of the approach we use. We also use a lot of data normalization tools, data cleansing tools, and because we operate in all three pillars, growth, business transformation, and employee experience.
We are well aware when we work with other customers because we do this for a living. When we work with other customers, their data is also not perfect.
Think about onboarding a building that's trying to connect all of its equipment to the cloud through Brainbox. That is not an easy task. But that doesn't mean we just throw our arms up and say, imperfect data. So machine. So the whole process has to stop. I use the same approach when we apply business transformation. We have to be pragmatic and just fixing every problem of the past is not going to define the future.
[00:24:43] Speaker B: For folks who are still grappling with the decision, was there something that kind of changed your thinking or is a natural given your journey, is that a natural way your thinking has evolved or what is advice you could give to people who are waiting for that perfect data or trying to get to a place where data is more easily accessible before they start their own AI journey?
[00:25:04] Speaker C: I would say the most important thing is to get started and to pick a few areas and to get started in those few areas. I think the big mistake, I see a lot of companies making this and I was at a recent event with peers and we were just chatting.
The big mistake companies make is they crowdsource and they try to have workshops and say, let's figure out ways to use AI. It sounds very intuitive, but the problem with that approach is it leads to so many ideas and you're chasing so many things, many of which have dubious paybacks.
And then you get into this false narrative of trying to quantify payback when there is none. And that leads to a deflation in the intensity of the effort. I think a better approach is pick a couple of meaty areas, go at them with intensity, drive some, drive some real roi. Don't worry about the data issue. The data issue can be solved as you go along.
And once you get some success, it kind of builds momentum and that becomes a good stepping stone to driving AI success.
[00:26:09] Speaker B: That's wonderful.
I think, Riyas, coming to your own career journey, right? You trained as a mechanical engineer, then went on to build this great career in enterprise software, IoT.
And now you've come full circle, right? As we discussed, back to the physical systems in the world of engineering. How has that experience working at Cisco, working at infor, working at SAP, lended itself to this journey? Now that you're on, on train, what are things that you have brought and you feel extremely glad that you have had that career trajectory and where are things where you're still discovering that, hey, I had forgotten that there were things like this which, which are helpful for a mechanical engineer.
[00:26:50] Speaker C: I was always fascinated with machines, even as a kid, even as an adult. I love aircraft, all types, not just commercial, but military aircraft. I love cars of all types, racing and otherwise.
I've always liked machines. I've enjoyed machines, which is why I studied mechanical engineering.
I remember studying the thermodynamic cycle, which fascinated me. You know, Adiabatic functions and isothermal functions. It was just something I gravitated to naturally.
And I graduated in the 90s, and that was a boom time before the dot com bust and technology was taking off. And my father encouraged us to learn how to code when we were teenagers, and I started coding when I was in my teens and actually built software. So the combination of this fascination with machines and this fascination with software was what led to me joining tech companies like SAP and Cisco. And it was great because I worked mainly with manufacturers at SAP and at Cisco, and I learned a lot about how to use this power of technology to take manufacturing to the next level, because I've always believed that a strong industrial base also leads to a strong nation.
And that's been a core belief of mine and been kind of reinforced through my career.
And Then, of course, IoT came along about 15 years ago, and that was the rage. And IoT is about understanding how machines operate without actually being at the machine and using the power of IP to track a machine's performance from afar, from remote, and using that to drive conclusions and to actually take action. And I had the pleasure of working at Cisco. We made a big acquisition of a company called Jasper Technologies, built a digital fabric that allowed literally millions of cars around the United States and the world to run. And that taught me a lot about the automotive industry, but also about digital fabric. So it was a natural extension. When Trane called, and again, I fell in love with what I had always studied, which is the thermodynamics, the vapor compression cycle.
And if you think about the opportunity, Mohit, 70% of the buildings in the United States have central H Vac in the Eastern Hemisphere, especially in Asia, in India and China, only 30% have central H vac.
And because of the way climate is changing globally, there's going to be a huge demand for H Vac. So I think the future for H Vac is very bright.
Yeah.
[00:29:25] Speaker B: I think if you think about your journey starting at Trane Technologies, right from it's a career shift, coming back to the titanium economy, what was the thing that surprised you the most?
[00:29:39] Speaker C: Just the scale of operation, the magnificence of these machines. I mean, I knew they were great, but standing next to them, you feel very small and you feel very excited at the same time.
[00:29:50] Speaker B: Got it.
[00:29:52] Speaker C: I've stood next to chillers that are literally the size of a school bus, and some of them are the size of two school buses.
And I love bending metal. I've always enjoyed being in manufacturing facilities and manufacturing plants.
And that was a very pleasant surprise.
[00:30:07] Speaker B: Got it.
[00:30:08] Speaker C: The other surprise was the culture of the company. I think the company has a tremendous culture. People have been here a long time.
My own boss, our CEO, has been with the company 40 years, if you can believe it. He started as a, as an adp, which is our early development program. A lot of my colleagues have been here over 20, 25 years. And that's pretty special. In the tech world, you don't see that as much. So coming here and seeing that kind of loyalty, that was endearing.
[00:30:33] Speaker B: Yeah, actually that leads to an interesting segue. Right, so Trane Technologies being in the H Vac segment, right, you guys are leading on AI implementation. You talked about how you are using AI within and for growth as well.
How do you compete for AI talent, for digital talent, with the tech companies of the world where you have before? Like, how do you get them excited to support you, to join you on this journey?
[00:31:01] Speaker C: At Trane, you know, there's two ways. Number one, you try to recruit people for whom mission matters.
And our mission is very clear.
We want to innovate for a sustainable future, for a sustainable world.
There are a lot of people, especially in the millennial and the Gen Z generations, who truly care about this.
And I shouldn't throw my Gen X under the bus. I think there are a lot of Gen Xers who care about this as well.
I think you should hire people who care about the mission. As you drove in today, you probably noticed that the street was named Sustainability Web.
You know, some people notice it, others don't. Hire the ones who noticed it because to them it matters. Right?
And if you hire people that are mission driven, they will bring their skill sets to us. And that's been my experience. I've hired some amazing talent over the past three years. We've been able to attract, when we post jobs, we get some of the best folks from some of the top universities, top companies, including from some of the best tech firms that want to join. And, and many of them talk about being mission driven and being inspired by our mission. So that's vector number one. Vector number two, you gotta be opportunistic as well. I mean, if you look at what's happening in the tech world these days, I say this slightly tongue in cheek. There's a lot of talent out there that wants gainful employment and that wants purpose. And then they come and try us and they say, hey, this is a wonderful company. This is a great place to work. I can see myself building a future here. And everybody here Seems to enjoy what they do and they've been here a long time.
That's exciting too. So those are the two kind of vectors that we use. But being mission driven, I think attracts a lot of people.
[00:32:40] Speaker B: All right, so let's look at that mission and look at the future. Right, so trained technologies, digital platform scales as you have envisioned it.
What does the commercial building of the future look like? And what role is trained technologies playing in that change?
[00:32:57] Speaker C: So we believe we will define this future as strain technologies. The future of buildings is smart, resilient buildings, buildings that are optimizing the use of energy.
Our CEO often says, based on industry facts, that most buildings are wasting about 30% of the energy that they pay for.
So let's say you pay for 100 units of energy, you're wasting 30 units.
Now, some say it's maybe closer to 40, but let's even take 30. That's a lot of wastage of energy. So how do we make these buildings smarter? How do we help buildings get more resilient? And we think this is a combination of things like artificial intelligence that we spoke about, cloud based bms, so building management systems that are cloud based and combined with artificial intelligence to drive unique value using things like hybrid dual fuel capabilities. So you're so using time of use to decide when to use fossil, when to use the grid, using things like battery storage to charge the battery when rates are low and discharge the battery when rates are high. So you're bringing down your peak, integrating with things like thermal storage. So using ice tanks, as you drove in today, you may have noticed some ice tanks. That's a way for companies to also reduce their peak.
And then of course, integrating with all kinds of all manners of on site generation, whether it's solar or it's wind or it's geothermal, the building of the future will be able to leverage all of these different options, almost like a portfolio of choices.
Overlay that with the rate structure of the grid and actually bring its peak down.
Now, if you do this at small scale, it's great for the individual building, but imagine the impact, Mohan, if you could do this at the scale of a small city or a big city like New York city, imagine the impact it could have on the resiliency of the grid itself.
So I believe there's a micro benefit at the individual building level, but there's also macro benefit at the national level for the grid. So we believe instead of predicting the future, we should play a role in defining.
[00:35:11] Speaker B: Yeah, that's excellent.
I think you threw out an interesting number out in there. Right. So you mentioned that 30% of the energy that a building is consuming is actually being wasted.
I think one of the levers that we talked about that, hey, AI can help reduce that waste. I think there is some skepticism on how much energy does AI itself take to produce these results. Right. Have you guys done kind of that analysis for yourself to figure out are we having a net benefit or how do we think about this towards a net zero emission aim?
[00:35:46] Speaker C: You know, my colleague Lee and I, last year, as we were preparing for a debate on the climate, we did some research and it's very interesting and I don't remember the numbers exactly, but we looked at how much energy was consumed globally.
It was about 60 terawatt hours globally. If you look at, we went to the EPA and got some numbers and we said, how much energy is consumed globally and how much of that is used by H vac? About 10%.
How much of that can be saved using AI? About 30%. So you end up, you do the math, you end up with a number. Let's call it X.
How much energy do all data centers, not just AI data centers, but all data centers use?
Let's call that number Y. So the amount of energy you save using AI on H Vac is X potential.
And the total amount used by all data centers is Y. X is about four times the size of Y, which means if deployed properly and completely. That's a big if. But if deployed Properly and completely, AI could create 4 times more energy savings just from H vac compared to what it uses.
It's an execution issue, it's not a technology issue, it's an issue of will. Right?
And I think as we define the future, we have an opportunity to make a dent. So I definitely think technologies like Brainbox, like training, AI control, will help solve the problem AI creates.
[00:37:19] Speaker B: And to your earlier point, there's a network effect to this, right? Because you think about it at a building level, but then at a grid level and then a city and a major city level. So I think those effects compound over time.
All right, so I think my final question, Riaz, right. In your role as the cdo, it's not a settled playbook in industrials, right? Every company has its own unique systems, its own unique requirements.
What do you wish you had known before the start about this role? And what does the CDO role for the next decade look like for you?
[00:37:55] Speaker C: You're absolutely right in saying it is not a settled role, but I see that as an opportunity because you get to define it.
I'm the first Chief Digital Officer for Trane Technologies and so I have a responsibility and an opportunity to define what that role looks like.
I believe in industrials, CDOs have to be outward looking, they have to be looking at the market, they have to be looking at revenue growth and they have to be focused on top line growth and on driving margin expansion for all the reasons we've discussed. There's huge opportunity to drive and create new services, there's huge opportunity to reduce the cost to serve. And through that there's a huge opportunity for CDOs to truly be growth vectors for their respective companies.
CDOs can also play a role in helping make companies more profitable internally. But I think the real value of a CDO is to look outside and to figure out how the company can be more profitable and more competitive at that stage.
In terms of what I wish I had known, I knew Trane was big, I knew the opportunity was huge.
But if you look at how our company has performed, when I joined the company our stock was 190 and a couple of weeks ago we crossed 500. So the company has done really well, the market has responded really well and I think customers are voting with their wallets and there's a tremendous momentum we have as a corporation and I'm very excited to be part of that.
[00:39:25] Speaker B: Any last words of advice for either CDOs who are beginning down this journey in industrials or CEOs looking to hire CDOs on how should they be thinking about this new role and what should be the remit?
[00:39:40] Speaker C: My simple advice is less analysis, more execution, because analysis often leads to paralysis.
And I've seen it everywhere. I've seen decks that are 50 to 200 pages that actually end up confusing my peers. And I chat with many of them on different councils and so on, and it actually paralyzes them from taking any action. I think a better way is to move quick, learn from both your success and your lack of success, iterate and keep moving. And I think that approach leads to more success and it creates a drumbeat inside of trained technologies. We call it robot Rolling Thunder every quarter, like thunder, it rolls right Every quarter. We bring out three things that are new, exciting and solve a customer problem. And we call these out in advance, we call them out for the next four or five quarters and then we deliver them quarter by quarter. So there's a sense of urgency and there's a sense of rapid and continuous achievements that are leading us towards a bigger goal as opposed to this grand strategy that will eventually get us there if all data was perfect and everything else aligned.
[00:40:56] Speaker B: Right.
[00:40:56] Speaker C: So I would always my advice would be get started.
[00:41:01] Speaker B: Got it. Thank you so much Riyaz for that very wide ranging and very honest interview.
I think for me personally the mission statement for Trane obviously is already inspiring but I think the vision you have set and the bias towards action is also inspiring. Right. I think there's a lot of takeaways for folks like me and obviously others listening out there on how do you actually get started down this journey. It's about taking the first step and not keep thinking and waiting for the perfect conditions. But yeah, no it's I think been great meeting you and knowing more about your journey and learning more from you. Thank you so much for this time.
[00:41:36] Speaker C: Thank you.
[00:41:42] Speaker A: Thanks for listening to INA Insights. Please visit INA AI for more podcasts, publications and events on developments shaping the industrial and industrial technology sector.
