Satya Nadella: Hannover Messe 2016

Remarks by Satya Nadella, chief executive officer, in Hannover, Germany. on April 24, 2016.


SATYA NADELLA:  Thank you very much.  It’s a real privilege to have a chance this afternoon to be able to be here in Hannover and have a chance to talk with you.

I’ve had a great opportunity in the last year to travel all over the world and meet with many businesses, many government leaders, small businesses, large businesses, and really reflect on our mission as a company.

Because at Microsoft, for us, the idea of empowering every company and every organization and every person on the planet to achieve more is central to everything that we do.

And reflecting on that, digital technology today, of course, plays such a central role.  Central role in our lives, in our organizations and how they thrive, and in our society at large.

And talking about that, I would say industry or manufacturing and digital technology is, obviously, very central.

If you look at the relationship between information technology and manufacturing, there’s nothing new about it.  This has been something that has gone on from a long time.  In fact, some of the first systems that got deployed in terms of information technology were because of manufacturing.

You can go all the way back to MRP or ERP or supply chain or even CRM applications were all developed because of the need of the manufacturing industry.

So the manufacturing industry, in some sense, drove information technology as much as anything else.  Even the embedded systems world of information technology was driven by manufacturing.

And so the question, then, is with Industry 4.0, what is that new inflection point?  What is new?

I would posit that what’s new today is that the very thing that you produce, the very thing that you manufacture, for the first time, is connected with all of the web of activity around it.  That is what’s really different.

Now, it’s not just the connection with everything, it is the ability to reason about that activity, that data that’s being generated continuously.

And not just, in fact, reason about it, but to gain predictive power that then can, in fact, be fed back to the operation of the thing that you manufacture.

These new digital feedback loops that I refer to as “systems of intelligence” is the new inflection point that we collectively across the software industry or the digital industry and the manufacturing industry are, in fact, bringing forth.

These systems of intelligence, in fact, don’t sit in isolation.  These feedback loops, the bringing together of IT and OT, as I like to think of it, as the convergence of information technology and operations technology is pervasive.  It changes how you will engage with your customers, how you, in fact, empower your own employees inside every one of your organizations to be able to gain insight from big data, take action from big data, and to optimize your operations and change the very nature of the business models around your industrial products.

These systems of intelligence are no longer just products you consume.  These are, in fact, systems of intelligence that each one of you as leaders in industry and manufacturing will build.

In other words, you are yourselves digital companies.  You’re first-class digital companies.  You’re not just manufacturing companies, but you are going to be world-class manufacturing companies with world-class digital capabilities.

And, to me, that is really how I think about Microsoft’s role.  In fact, there was a lot of talk about is there a separation between what is industrial and what is digital?  I think going forward, really, there isn’t.  If anything, these two things have to harmonize for driving growth in our respective companies and in our respective industries and economies.  And that’s what we want to enable.

I want to illustrate this point, these systems of intelligence, by talking about, perhaps, four examples.

Let me, in fact, start with Siemens.  We’ve worked with Siemens for many years.  And one of the fascinating things that I see happening is in the healthcare side of Siemens, they of course produce these world-class machines, whether they be CT scans, ultrasound machines.  All these machines are connected today.

But more than the fact that they’re connected, what they have done is they’ve built a collaboration cloud software and service, in fact a cloud platform, for doctors, radiologists, patients to be able to collaborate on all of the connected information.

And something like 200,000 patients every hour are being seen by Siemens machines.  So just imagine if now all this data is being used to reason over patient outcomes, change the dosage that anyone can get exposed to, and really have that kind of impact.  That can be pretty profound.

So this system of intelligence, this collaboration platform is really a Siemens product.  Of course it’s built on some of the underlying platform capabilities we provide, but I would posit that what Siemens has done is bridge that gap between what is an industrial company and what is a digital company by providing, in fact, their own digital products.

Another example is what’s happening with Liebherr.  They produce industrial refrigerators.  Now, I’ve learned a lot about industrial refrigerators recently, and it turns out that they need these very high-precision capabilities.

In Germany itself, for example, 26 million drugs are sold every year which are required to be managed within a very tight band of temperature, something like 2 degrees Centigrade to 8 degrees Centigrade.

Now, in order to be able to manage something in such a tight bandwidth, that kind of precision, you need a digital system.  So that’s what they’ve done.  They’ve built their own system of intelligence which allows every industrial refrigerator of theirs to be continuously monitored so that it can, in fact, produce that right temperature and manage that right temperature.

And if anything goes wrong, they see it before it goes wrong and they can, in fact, roll a new machine and, in fact, have an SLA around it.

So this change of business model and creation of this system of intelligence is fundamentally transforming even the refrigeration industry and then in many, many sectors like healthcare.

Another example is Jabil, which is an American contract manufacturing and design house that has many factories around the world that they run.  And they’ve always had world-class shop floor automation and factory floor automation.

But what they’ve gone and done is the next level where the entire factory floor, the entire production line is itself an adaptive learning system.  So that means even if there is a mistake made in the first step of product, they’re able to connect back to the cloud, use machine learning, detect that mistake, and correct it before it goes all the way to the end of the production line.

Just imagine now.  The idea that you can detect this because you are able to use compute power in the cloud to run your production floor operations more efficiently by itself is revolutionary.

But the more interesting thing to me is that it’s a continuous learning system.  So that means every operation and every run, you are becoming better and better.  The system of intelligence is not a static thing, but it’s a continuously improving digital feedback loop that many, many people can benefit from.

And another example is Rolls-Royce.  Now, Rolls-Royce makes these aircraft engines.  They have something like 13,000 aircraft engines at any given time that are in flight.  These all have sensors.  They collect terabytes of data from these flights continuously, again, looking at them, monitoring them.

And the most interesting thing is that they have created a software product, a digital product, by taking the data from their aircraft performance and conflating it with many other forms of information and turning that back into airlines so that they can have an efficiency service.

So you can just look at this.  I mean, basically, Rolls-Royce is going from being essentially a manufacturer of great, world-class aircraft engines to building even software products that are going to not only drive better performance of their engines, but also drive additional value to their airline customers when it comes to how they operate.

To give you a little bit of a flavor of how this works, I want to introduce one of my colleagues, Tim, up on stage to just give you a flavor for what’s possible now with aircraft engines.  Tim?

TIM:  Good afternoon.  Several decades ago, Rolls-Royce pioneered the “power by the hour” service model, transforming their business to effectively sell air time as opposed to just engines.

As Satya mentioned, now they have 13,000 engines in operation all around the world, constantly pushing out gigabytes of data, especially on long-haul flights.

So today I’m going to show you a vision of how Microsoft and Rolls-Royce are working together to transform their business again, leveraging these systems of intelligence to make better engines and a stronger customer service model.

Unplanned delays and unscheduled maintenance can cost up to $1 million a day for an airline in trickle-down costs, not to mention delays to customers like us trying to get from A to B.

So for this reason, in this dashboard, I can see an entire summary of my fleet, its health status, where my planes are, and actually where they’re going.

I can see critical KPIs like on-time performance, schedule stress, and freight utilization, all of which are great lead indicators for a potential disruption in my operation.

I can see flight schedules, maintenance schedules, and even the maintenance capabilities I have in different locations and airports.

The early watch list escalates advisories on particular aircraft that require my attention.  For example, this one in Frankfurt at the top on the left.  Let’s take a closer look.

Aircraft and engines are incredibly complicated machines, but they’re also very well monitored.  Here, you can see a summary of all the critical systems on this aircraft.  I can see it’s on the ground in Frankfurt for about several more hours, but it’s not due for service for quite some time.

I can also see on the right the remaining useful life of both my engines, and the minimum and recommended fuel levels that help the crew decide how much fuel they should take for the next flight.

I can also see previous flights and the variance from the planned route and planned altitude.  Even a small deviation can equal a large increase in fuel consumption.

I see this issue with the propulsion system.  Let’s take a closer look at that.

Here, I can see one of the engines that’s attached to this aircraft.  I can see all the service history and service notes from technicians who have worked on this piece of equipment.  I can see all the information I need in order to make great decisions about engine wash optimization, and where and when I do the next scheduled maintenance.

As you can imagine, pulling an engine or an aircraft out of service means it’s not available to me to fly a route.  And if that’s in an inconvenient location where I don’t have time or experienced technicians or inventory, that’s incredibly inconvenient.

So here I can see that the primary fuel pump is reaching the end of its useful life.  And while the aircraft could safely continue for the next few flights, it might make sense for me to fix this part now rather than risk having a potential disruption down the line.

Here I see the results of my complex analytics, all shown to me in an embedded Power BI chart.  I can see the compounding and contextual factors that are all summed up together to conspire to make this part prematurely reach the end of its useful life.

I can see the pressure through the pump into engine one is dropping, fuel flow is decreasing, and the power through the pump is much less than we’d expect it to be at this stage of its life.

I can see that this aircraft is not due for scheduled maintenance for quite some time.  But we do have time here in Frankfurt, we have the parts and inventory available and experienced technicians to take care of this issue right now.

We think this has happened because the plane looks to have been flying a lot of short-haul routes.  Short-haul routes are very hard on the engine and systems because the engines don’t get to run at a consistent RPM.

Couple that with what looks like some pretty harsh operating environments, and that’s what’s led this pump to fail prematurely.

I’m going to request this maintenance be taken care of now.  We’ve got time at the gate, we’ve got the people and the parts.  I’m going to execute this workflow, have this work taken care of right now, and get this plane on its way to the next destination safely and on time.

Here you’ve seen how Rolls-Royce and Microsoft are working together to once more transform their business.  Utilizing the intelligent cloud and systems of intelligence to build better engines and a better customer service model.  Thank you very much.  (Applause.)

SATYA NADELLA:  Thank you, Tim.

The idea behind that was not as much to talk about the technology that we produce, but it is to be able to point out that at this point, Rolls-Royce is building a software product as the core value of what they do.

This is what we envision for every industrial company.  We want to be able to create these platforms that can create, in turn, more digital technology from every industrial company.  And that’s really the essence of what we see happening all over.

But now the question is:  What is next?  What do we imagine is going to happen that’s going to, in fact, have another profound impact?

Our industry of digital technology has been punctuated by these category creation moments where we have fundamentally changed how humans input and output digital technology.

And we are really at the forefront of one such change:  Mixed reality.  In fact, it’s for the first time where our field of view becomes an infinite display that can not only have the analog world that we see, but can, in fact, have digital artifacts superimposed.

The fact that we can mix analog and digital worlds is going to have a fundamental impact.  This is not virtual reality, this is mixed reality.  And mixed reality will change industrial design.

I mean, just imagine, you’re building a car, an elevator, or any part.  You’re going to be able to not only see the 2-D of it inside the computer, but you can, in fact, see the holographic output of your design program right inside your room.

So industrial design is going to change.  How we learn and teach machine learning engineering will change profoundly in terms of how we can go about making the skills of engineering much more widespread.

Worker safety inside of an industrial production line, where you have a lot of automated robots; just imagine knowing the radius of a robot so that it really improves the industrial safety of our workers.

And, in fact, the field service.  You can now have people with HoloLens in a remote location undertaking the most sophisticated repair operations because they can tap into the experts wherever they are who can, in fact, have holographic output being superimposed on the repair operation that they’re doing.

So think about how this new medium of computing can, in fact, help industrial companies change their own operation.

We’ve partnered, in fact, with Japan Airlines to do that very scenario of a field service technician who can now make sure that they’re able to do these sophisticated repair operations that they previously had to get a specialist in because of the holographic nature of communications between the expert and the remote technician.

So let’s roll the video to just give you a flavor for some of the scenarios being enabled by this new computing paradigm in HoloLens.

(Video segment:  HoloLens.)

SATYA NADELLA:  I really love that line when it says, “When you change the way you see the world, you change the world you see.”  Because I think that that’s fundamental to how this new paradigm can, perhaps, have an impact in all industries and in all walks of life.

I wanted to close, though, by talking about the role of technology more broadly.

The fundamental fact is that we, whether we be in digital technology or industrial technology, we do not live in isolation.  We live as part of a global economy and a global society.  And I think many speakers before me referenced even Society 4.0 as opposed to just Industry 4.0.

And if that is what we really need to aspire to, a new society, then we as private enterprises, as multinational companies, have to look to all the constituents that we work, whether it be the citizens inside of any given country, nonprofit organizations, and the broader economy, and really look to how are we making the environment more sustainable, the economy more viable, and people more skilled at the jobs of the future?

And to really reflect that and take our mission and apply it more broadly, we have an initiative called Public Cloud for Public Good where over the next three years we have put $1 billion to work to really bring that same compute power, the same capability of having systems of intelligence, really helping all walks of life.  Whether it be in education, whether it be in teaching computational skills, or even in healthcare.  In fact, we’re working with the University Medical Institute in Hamburg to be able to provide them the same capabilities that are now being used by many industrial companies to understand immunology and genomic sequencing to treat cancer.

So the idea that the technology that we have with us today can be broadly used to detect school college dropouts, and have the taxpayer dollars be used more wisely by the state to promote students staying in school and make outcomes of education better or health outcomes better, are as important for us to have an impact on as anything that we do as businesses.

So I just wanted to close with that and thank you again for the opportunity to share with you my thoughts today.  Thank you very much.