June 4, 2018 | Written by: John McDonald
Categorized: Continuous Engineering
Share this post:
In part 1 of my article, I shared with you a vision of our world from the vantage point of systems and automation. In part 2 of the series, I will explore how we as individuals and organizations can really take advantage of the new data-driven economy. Don’t get left behind: ensure your organization has a seat at the table.
The internet of people is ending
My body has five senses. My iPhone has twelve senses – most of which are designed to make me smarter. These “senses” are made to augment my own, working with me to compliment my existence as a living data source. On its own, without any human input, a smartphone isn’t very smart at all. What’s the Apple health app without data? What is Google Maps with no movement? When I flew from Indiana to Munich to speak at the IBM CE Summit, the first thing my phone did when I landed was to reset its internal clock. Without being connected to the internet, it can’t even keep time well.
And, although we say we carry around all of human knowledge in our pocket, the truth is there’s very little human knowledge on our physical phone. It’s just a gateway to make me a smarter endpoint to the Internet. It’s not really information in our pocket, it’s a connection point. The Internet is what makes us smarter.
An absolute explosion of software and everyday devices
The F-22 Raptor, a device largely conceived of in the late 1970s, has 1.7M lines of software code – which was a significant amount of software at that time. Fast forward only slightly to the mid-Eighties – when the Boeing 787 Dreamliner was conceived, containing 6.5 million lines of software code – a very sophisticated device.
However, the 2018 Mercedes Benz S Class 550 has 20 million lines of software code in it – 14 million of which are in the radio – making it the most sophisticated computing device that humans can interact with daily.
What does all that code enable? It enables data to be sent from that device.
The dawn of the data-driven economy
But what’s happening is a shift is now occurring with how we are sending that data and what we’re using it for. The early era of computing – marked by mainframe computers between the 1960s and the 1980s – was marked by very large central computers that controlled all the data and applications. We connected to them with what was called a dumb terminal. A dumb terminal had no actual function other than to take in a keystroke and send it to the mainframe and display a character generated by that mainframe.
Starting in about 1980, we started to shift all that power out to the edge in the form of PCs and then laptops. But to be fair, there were some challenges we never really solved when that happened. Things like, how do I know that you have the right data? How do I know that you should see that day? How can I keep you from copying that data onto a floppy disk and carrying it out of my business?
Shift and twist
In 2000, we started to address these problems when we recentralized all the applications and data in the cloud. Now, the new form of the dumb terminals is a mobile device, which pretty much brings us “back to the future” – with a twist.
Remember that car radio with its 14.7M lines of software code? What’s happening is we are redistributing the data and the process power out to the edge – just like we did in the client/server era. That means that the role of the cloud changes from the place where all of the data and connection goes, to a place where we select information to go to be interpreted and understood. A major shift is once again occurring in the landscape of how computing works. What this represents is the fourth major baseline shift in how our global economy is organized.
The first three phases of our global organization structure
The first major organization structure of our global economy was agrarian – organized farming. You farm so I don’t have to farm. Prior to that, everybody had to farm in order to eat. Suddenly, we had farmers that farmed so I could buy their products in the market place.
The second one was the manufacturing economy – or the industrial revolution – making things. You go make things in the factory so I don’t have to. During the nomad era, people had to make everything –the bowl they ate from, the tools they needed to build the home they lived in, weapons to find food. All these things had to be made by the person who intended to use them. In the medieval period, we realized the power of specialization, creating a guild and barter-based economy: you make what you are good at and make a lot of it, and I will reward your skill with my own product. Also part of the industrial revolution – the first time mass production enabled mass distribution – drastically reducing the cost of goods and providing greater access.
The third major organization structure – our global economy – happened relatively recently. It’s the revolution of how things move across the planet. This shift happened shortly after World War II, after the rise of container ships, jet air transport, and interstate highways. We have continued to find more efficient and effective ways to move things on a global scale.
When I ordered my iPhone, I got a shipping notice from the factory in China. I could track my device as it moved closer to me – via planes, trains, automobiles and boats – all the way across the planet to my front door in Indiana.
Data is now the great creator and destroyer of business value
Here comes the fourth major organization structure of our economy and it’s based on data. Let me illustrate this with some examples of what I mean by a data-driven economy.
- The world’s biggest hotel company is Airbnb and they don’t own any hotels.
- The world’s largest retailer is Alibaba and they don’t own any stores.
- The world’s largest car rental company is Uber and they don’t own any cars.
This is not a future statement. It’s not just happening. It’s already occurred. These companies are the leaders in their categories and they own none of the traditional assets that their predecessors had to own to be a player in that category, let alone become the market leader.
Photo: John McDonald, CEO, ClearObject
So, what do these organizations own? They own data. Uber knows very little about you. They know that you’re on a street corner and they know you need to get from here to there; they know your credit card number. In this new economic environment, with that very small amount of data, Uber can build a multi billion dollar business.
Here are three enterprise examples where this new organizational structure of our data-driven economy has already taken root:
Rolls-Royce has pioneered a business model called power-by-the-hour. In this model Delta Airlines and other airlines in the world are phasing out owning jet engines. Instead what they do is they pay Rolls-Royce for “power-by-the-hour.” They pay a recurring fee to Rolls-Royce to supply them with jet engine service. What this means is that it’s Rolls-Royce responsibility to make sure that there are two working engines strapped to that plane before it pulls back from the gate. If they have to put fourteen on there until they find two good ones, that’s on Rolls-Royce, not Delta Airlines.
This is great for Delta because it means it’s a recurring expense that they can budget around. It’s also really good for Rolls-Royce because it creates additional revenue streams. Anybody in the technology business will say this is a wonderful thing. What it means is that the data coming from the engine is immensely important. It’s very important from a predictive maintenance and quality perspective because it can be the difference that month between Rolls-Royce having a profit or a loss regarding the service they provide.
The data coming from the engine is more important than the engine itself. It’s not a far stretch to think that perhaps Rolls-Royce could start doing power-by-the-hour for engines that they didn’t even make. In fact, it’s not even too much of a stretch of the imagination to consider, if they got really good at it, that Rolls-Royce could phase out the production of engines and only do maintenance and support of engines that were produced by other companies.
Cummins Engine make diesel engines for various large-work trucks. These engines can become very finicky regarding their emissions. They can very easily get out of whack. And if they get too far out of whack, various government agencies make them do what’s called dual-rate the engines – which is to reduce its power. If you’re a fleet operator, you want to know when something happens to your vehicles – quickly.
Fault codes turn on the check engine light into car. In a Cummins Engine, the moment an engine system fault occurs, the telematics system instantly transmits key engine system and GPS data through the telematics connection. Whenever one of these fault codes happens, to the tune of about 30,000 of them per use, these messages are generated and sent to the fleet operator within one minute – letting them know what that fault code means and what to do about it – pull over and stop, wait till the next service annual, etc. The Cummins diagnostics system can pre-order the required parts through their internal network, intercept the truck at the nearest Cummins Care Facility, and get it back on the road faster.
Cummins used to win new business based on traditional measures like torque. Now Cummins is winning business based on the data strength in this new capability it offers. Just like the Rolls-Royce example, you can imagine a world where Cummins Engine might provide this service to other engine manufacturers, a world in which they cease production of their own engines in favor of doing the data streams around them. What if one of Cummins’ competitors had developed this capability first? How long do you think Cummins would be in business if their main competitor was winning business based on the data stream rather than a stupid engine without this capability?
Data is the great creator and destroyer of all business value going forward. The data that comes from the device is far more important in the new data driven economy than the old economy based on making things.
Ever since I was a kid, I’ve been taught not to mow the grass in the bright hot sunshine because it’s bad for plant health. Now we’ve been farming in the bright sunlight of the day since we’ve been farming. Why? Because we’ve had to see to do the job. Yet, what we’re learning through early experimentation and autonomous agriculture is that it’s better for the plants if we farm at night.
In an automated world, we can deploy vehicles that roll across a field, stop, deploy sensors, understand the various levels of PH, different chemicals and water in the ground – in the dark. We can spray the appropriate amount of nutrients using only what’s needed, then move on to the next plot – 24 hours a day, seven days a week, 365 days a year – like an intelligent factory.
Right now, the farm field is a production factory without instrumentation. We take pictures of farm fields at the beginning of the season and after the season is over. This shows us the starting point and the result, while offering no context as to how we got to this point. These two data points don’t have the ability to enhance and evolve our farming practices, because they do not provide any evidence or context that could inform my future process.
What needs to change in the new data-driven economy?
Our perception of jobs and education
People will need to be retrained to make them productive participants in this new data economy. Let’s take the example of a truck driver. Yes, we will need fewer of the traditional kinds of truck drivers. We will also need a different sort of driver – one who can function as a ‘platoon’ leader. But, is the current education system in place up to the task?
Our current educational and economic climate is an early indicator that this generation of individuals understand that the data economy requires a different skill set than what is available through the current system. Think about all those truck drivers that need to be re-educated? Our education system is not designed to retrain tens of thousands or millions of people to become part of the data set.
Our ability to discern between real and perceived security threats is critical – but when it comes to trusting humans versus machines or systems, our inclination is place our trust where we can develop an interpersonal bond.
Let’s say you go to Amazon.com to order that shirt. You get to the last page of the ordering process and it asks you for your credit card. You start typing in your credit card number and you get that feeling in the pit of your stomach. You convince yourself it’s safe, and you continue typing and hit send. When you’re done, you go out, sit in a café, order lunch, because you interpret that feeling in your stomach as hunger, not fear. When the meal concludes, you find yourself handing your card over to a stranger – the waiter – to pay for your meal. The waiter walks away for five minutes with your physical card and you don’t think anything of it.
What’s different between these two scenarios is our perception. Humans naturally cannot trust computers, at least not yet. And although it’s fleeting, in the time it’s taken you to sit down to lunch and order a meal, you’ve developed an interpersonal human bond. Now, which of these two systems is more secure when it comes to handling your credit card? Amazon by far is more secure; however, your mind perceives it as precisely the opposite.
There are all kinds of technology available to ensure the security issue with Amazon is minimized, and if there is a breach, Amazon is held responsible for betraying our trust. But you can never create any amount of technology that covers a perceived security issue.
Food for thought
Consumer IoT is much further behind than industrial IoT because we are unwilling to relinquish control of our personal data stream. Take this scenario as an example:
You’re driving down the road in your car and your car notices that you’re not keeping your lane as effective as you did about 15 minutes ago. And it thinks that you might be tired because it’s three in the morning. In response, you hear a message on your car radio asking if you’d like a Caramel Macchiato with an extra shot of espresso. You see, your car radio knows that there’s a 24-hour Starbucks up the road. If you say yes, the radio orders the payment, sends payment information, orders the coffee and puts a map on the screen directing you to the drive through where you get your coffee.
In this scenario, the car has just saved its own life as well as yours. If any car that is less than five years old has a radio with 14 million lines of software code in it, and the ability to do what I’ve just described, then why doesn’t this happen now? It’s because we don’t want it to.
Think of it this way: Who else do you not want to know that you’re weaving in your lane at 3:00 AM? Your insurance carrier, your partner, any service station or hotel along the road that might be able to spam you? Do you really want to have ten messages sent to you offering you a room for half price if you check in within the next 30 minutes? It’s the same reason why we turn off location tracking in all our mobile apps.
In consumer-based IoT, individuals are not franchised to participate in the control of their own data stream or the value it creates. If we received a check at the end of every month for the use of our data, we might be more inclined to giving other people access to it.
Overcome the complexity of connecting requirements, design, development and deployment. Discover how you can speed product development and take full advantage of the data-driven economy. Learn more now.
Don’t miss Part 1 in the series, Back to the future: Autonomous Vehicles.
About the author
John McDonald is the CEO of ClearObject, Inc. Magazine’s fastest-growing IT company in Indiana for 2014, 2015, 2016 and 2017, and winner of the 2016 Entrepreneur Magazine 360, the 2016 Deloitte Fast 500 in Technology, the 2016 CRN Next-Gen 250 and 2015 and 2016 IBM Beacon Awards, the highest honor given to a business partner.