Driving Sustainability with Connected Intelligence
Rob peels back the onion on this powerful technology to show you what they can do and how they can deliver value to your organization.
Once you’ve added telemetry, virtual, static, or command properties, it’s time to make use of this #metadata with rules.
The first steps in deriving value from streaming telemetry data revolves around pattern matching, key performance indicators (KPIs), & filtering. You therefore need to specify one or more rules to be associated with each “telemetry property” you’ve defined in your digital twin model. This is accomplished through the use of simple operators such as equals, not equals, greater than, greater than or equal to, less than, & less than or equal to.
Let’s say you’ve defined a “telemetry property” for the “left front tire pressure” of a car digital twin model with an “integer” data type and “PSI” unit of measure. To create a “green” KPI between 30 & 35 PSI, you would define a rule looking for values that are >= 30 & <= 35. Using simple IFTTT algorithms, the event processing engine of your Edge or Core IoT platform would apply those rules to incoming data & trigger an action for values outside that range.
In my last two posts about defining Digital Twin models for your IoT system, I described “telemetry properties” that map 1:1 with device sensors and their associated data types and units of measure. I also described “virtual properties” that are a sort of “soft sensor” whose value is derived through a calculation from sensor values and other data sources.
Today I’m adding “static properties” to your Digital Twin model which are values that typically don’t change. If I use a car as an example, static properties could be things like the length of the car, the number of cylinders, displacement of the engine, and the volume of room in the trunk. Static properties are necessary to have a more complete view of the actual entity (car in this case) and to be used as reference data for analytics when defining the Digital Twin model that instances of Digital Twins will be derived from.
A digital twin model is used to define a class, or type, of entity/thing. In other words, you define all aspects of a type of thing just once, rather than defining it over and over again for each individual thing. Each instance of a digital twin derived from a digital twin model will inherit its attributes. The digital twin model tells the event processing engine in your Internet of Things platform what to expect by providing “telemetry properties” that include the data labels, data types, and units of measure for the incoming data to assist with pattern matching. No matter what level of analytics or machine learning you’re using, you won’t be able to derive any actionable insights without knowing the details of the data that’s arriving from your IoT endpoints.
The digital twin model indispensable.
Moderated by WIRED’s Editor-in-Chief, Nick Thompson, this lively debate across multiple technology disciplines was arguably the best panel I’ve ever served on. There was never a dull moment for the audience as we painted a picture of what the 4th Industrial Revolution will look like.
Watch the video below:
Panelists:
Finding ways to increase the utilization of these often-idle assets is one of the biggest payoffs for an organization embarking on the IoT journey.
Underutilized assets like cars, offices, MRI machines, assembly lines, dishwashers, partially empty trucks on the road, conference rooms, & thousands of others must have their utilization optimized so they can earn their keep.
You wouldn’t let employees spend 75% of their working time taking smoke breaks. Your machines & supply chains should be no different. These underutilized assets & processes must be put to work in a full-time capacity.
This phenomenon is reminiscent of previously-idle servers in data centers that had their utilization boosted via operating system virtualization. They were put to work so they could earn their keep.
Connecting assets to IoT systems & analytics in order to reduce unplanned downtime & increase their remaining useful life are important first steps on your IoT journey to reduce operating expenses.
The next step is to increase the incremental utilization of these assets by connecting them to IoT & Blockchain facilitated digital marketplaces where their spare utilization can be rented to others in order to boost revenues.
As Machine Learning gets better at forecasting the level of asset utilization, a futures market can be created where counter-parties can trade upcoming free utilization.
Share idle assembly lines in factories w/ partners. Share free hours of MRI machine time w/ patients at other hospitals & medical plans. Know where trucks are going & how full they are so we can fill them. Know where unused food & idle commercial kitchens are & feed the hungry.
To sum up: Connect your assets to keep them healthy, know their utilization, & reduce operating costs. Offer up the underutilized portions of your assets to digital marketplaces or other matchmaking engines to earn additional revenue & help others who need what you have.
Twitter is made up of people who have something to say. These people express themselves by Tweeting. Oftentimes, no one is listening. There are other people on Twitter who choose to follow those Tweeters in order to listen to what they have to say. Those people are called Followers. These folks often follow lots of Tweeters to understand the state of their collective minds. A Follower gets notified when a Tweeter they’re following says something. Through the clever use of Hashtags, followers can also choose to search for specific topics aggregated across many Tweeters in order to derive larger insights. Depending on the insight, the Follower takes action. Sometimes, a Follower wants to say something to a Tweeter. They can do this with a Direct Message (DM). Of course, a Follower can only send a DM if the Tweeter has authorized this by following the Follower back. The Follower may say something to the Tweeter that either changes her behavior or updates her state of mind.
You never know.
The Internet of Things is made up of machines that have something to say. These machines express themselves by Publishing their telemetry data to some nearby or far away computer system over a communications network. Oftentimes, no one is listening and that data just piles up. There are computers, people, apps, analytics, machine learning and automation systems who are interested in what the machines have to say. They are called Subscribers. They Subscribe to lots of Publishers in order to know the current state of their collective health or performance. A Subscriber often gets notified when a Publisher streams new data which allows them to process that information in near real-time. Subscribers can also choose to search through data swimming in a lake to derive larger insights. Depending on the insight, the Subscriber takes action. Sometimes, a Subscriber or some other endpoint wants to send data to a Publisher or group of Publishers. They can do this through a Command and Control channel. Of course, a Subscriber can only send a message if the Publisher has authorized this action. The Subscriber might send a Command that either changes the Publisher’s behavior or updates its configuration.
You never know.
I realize it’s easy to get overwhelmed with the sheer complexity of these systems that are transforming our world. That’s why it’s important to maintain the simplest view of what these IoT systems are actually doing. Explaining it to others gets easier which allows you to focus on the specific element that drives value.
Organizations will grapple with a variety of digital technologies and skillsets this year to become more data-driven in order to improve their agility and decision-making capabilities. As always, they’ll be looking for ways to simplify operations and get more done with less. We predict the concepts and trends listed below will light a path for organizations to show them the way forward:
The ongoing journey to move data, apps and other digital assets from private, on-premises data centers to public clouds will continue unabated as organizations look to reduce or eliminate internal ICT functions and responsibilities. Even in the midst of cutting costs, organizations will still struggle with concerns around cloud vendor lock-in via PaaS which will benefit IaaS virtual machines, container technologies like Docker and container orchestration technologies like Kubernetes, Docker Swarm, Mesos and Marathon. Overall, Amazon AWS plus Microsoft Azure and Office365 will continue to be the biggest beneficiaries of the public cloud megatrend. Along the way, one of the stair steps that remains on-premise is something called the Fog or the Edge. If you’re familiar with how content delivery network (CDN) proxy servers around the world cache and speed the delivery of Web content to your browser, Edge gateway devices do something similar. With more and more of an organization’s compute occurring in distant, public clouds, Edge devices residing on the local network can cache, aggregate, analyze and speed up cloud content to give employees inside the office a better experience. Edge devices can also be used with the Internet of Things where they connect to machines and cache, aggregate, and analyze data locally instead of waiting for that data to be transported to a distant cloud. Since neither people nor machines are vary tolerant of too much latency, expect the adoption of Edge gateway devices and associated local storage to surge in 2018.
As organizations reduce the number of digital assets and activities that take place in-house, the primary role of ICT departments will be to create and maintain fast, reliable connectivity via wired and wireless technologies. Wired networking will be “more of the same” as we push speeds forward with fiber optics and Gigabit Ethernet to shuttle employees out to the Internet. Wireless is where things get more interesting. Inside the office, organizations will continue rolling out 802.11ac Wi-Fi access points running in the 5 GHz band to deliver data and high-bandwidth content like HD video to any device. Outside, the 3GPP has officially signed off on the first 5G specification which promises to deliver greater bandwidth, lower latency, better coverage, lower battery consumption and a higher number of simultaneously connected devices. As you might imagine, it will take some time to roll out technology based on this spec so we will look to get more mileage out of 4G technologies like LTE Advanced. On the slower side of things, you have Low-Power, Wide-Area Network (LPWAN) technologies that are making great strides for certain Internet of Things use cases. The ability to create a large wireless network in places where no cellular coverage exits is compelling for organizations capable of managing such a system. If you have devices or machines that don’t send much data every day, require years of battery life, or need to send data over long distances, one of the many LPWAN technologies might be a good fit. Whether you’re inside or outside, looking for narrowband or broadband, there’s plenty of wireless choices for organizations in 2018.
While the mobile device revolution has been the biggest megatrend of this new century, the torch has now been passed to the Internet of Things. When you think about it, they’re not terribly different from each other except for the endpoints. Mobile device endpoints are proxies for people and Thing endpoints refer to machines (intelligent or otherwise). They’re both sending data about themselves and other topics of interest over a network. Both interact with apps, analytics and other on-prem or cloud data sources to derive value and business intelligence. In order to regain a level of simplicity and perhaps sanity, organizations will push back against the use of multiple enterprise platforms for Mobile people and IoT machines. Additionally, many organizations will wring their hands of having to understand an alphabet soup of protocols and myriad IoT standards and revert to using the same Web and Internet standards they already understand. Just like they currently do with Mobile and the Web, organizations will insist that IoT sends and receives JSON data to and from URLs over HTTP/REST while being displayed via HTML5, secured with TLS and brought to life with JavaScript. This use of familiar, widely-used, “good enough” Web technologies will win the day over the more advanced but esoteric technologies currently employed by IoT platforms. This move to simplicity and familiarity will reduce friction and help the Internet of Things deliver value and fulfill its promise the way the Mobile, Web and the Cloud have. Expect big changes in IoT for 2018 along with a big shakeout of the hundreds of Internet of Things platform companies.
The rise of Digital Twins will give every organization the starting point they’re looking for to begin their Digital Transformation. A Digital Twin is essentially a digital representation of a physical object. It can be a machine, a person, a complex mechanical subsystem, a collection of machines working together on an assembly line, or even a process. These twins have attributes or properties that describe them like a person’s heart rate or a motor’s temperature or current revolutions per minute (RPM). Organizations can assign key performance indicators (KPIs) to the current values of these properties. A red heart rate KPI might be 200 whereas a green motor temperature KPI might be 200 degrees Fahrenheit. Digital Twins can exhibit behavior by executing programming language and/or analytics code against the combination of their current property values and associated KPIs. Not only does this bring everything in an organization to life, it also facilitates the running of simulations to see how things will behave when different types of data points are fed to these Digital Twins. This is definitely the most promising and exciting technology for 2018.
Unrelenting cyberattacks keep organizations in a defensive posture rather than moving forward with important digital initiatives and deployments. While we won’t cover the myriad security steps every organization must follow in order to stay ahead of individual and state-sponsored hackers, this is one of the most important functions of an ICT department. Organizational leaders who don’t take this seriously by not funding the appropriate security technology or staffing the appropriate security employee headcount do so at their own peril. Needless to say, organizations must prioritize the privacy and protection of data, people (employees and customers), and systems if they want to remain viable. To turn up the heat a bit, the European Union’s General Data Protection Regulation (GDPR) becomes enforceable on May, 25 2018. This regulation gives control back to EU citizens and residents over their personal data by strengthening data protections for all individuals within the European Union as well as the export of personal data outside the EU. Quite a few companies operating in countries across the globe play it fast-and-loose with the security and privacy of individual data without user consent. This comes to an end in May when companies can be fined up to €20 million or 4% of their global annual revenue, whichever is greater, for violating this regulation. Any company operating in the EU must obtain explicit consent for all data collected from an individual as well as reason/purpose of using and processing that data. Additionally, that user consent may be withdrawn. Many companies around the world haven’t made the necessary changes to their digital systems to be compliant with GDPR and will be in for a rude awakening in 2018. Data privacy and security matters in a big way.
While data and analytics systems have been around for decades, the amount of data collected for analysis by organizations has increased exponentially. With a 50x growth rate from machines alone, the Internet of Things has become the newest data source for organizations to analyze. Lots of little data integrated from people, machines and business systems adds up to an overwhelming amount of Big Data to make sense of. Luckily, there are an increasing number of streaming and batch analytics systems and tools to tackle this job. Making this trend better is that most of these technologies are open source and free which helps level the playing field between small, mid-sized and large organizations with varying amounts of money to spend. Head over to Apache.org. Another interesting trend in data science is how Python has surpassed R as the most popular language for Machine Learning. An increase on online courseware, an abundance of scientific libraries, and the fact that Python is one of the easiest programming languages to learn, means you don’t always have to be a PhD in Statistics to get the job done. Virtually every organization in the world is looking for Machine Learning/Deep Learning expertise, so this trend should help the supply side of this equation. The last analytics trend that is coming on strong in 2018 has to do with where data is analyzed. It will no longer be the exclusive domain of the cloud or large clusters of servers. The need to answer questions and make decisions more quickly is driving analytics of all types out to the Edge. Thanks to Moore’s Law and the need to eliminate latency, more and more edge gateway devices will be performing IFTTT and even Machine Learning predictions (with models trained in the cloud). There’s no shortage of important trends that are simplifying advanced analytics for organizations in 2018.
Clearly, 2018 is going to be a transformational year where properly-equipped decision-makers and leaders can shift their organization into the next gear to accelerate their digital transformation. Hold on tight.
This was my first startup after getting out of the military. We brought vending machines to life with embedded software and hardware, early wireless data technology, and graphical software running on Windows.
Our IoT technology allowed operators to know the current state of their vending machines from across town or even across the country. This revolutionary service reduced costs, increased sales and enhanced customer satisfaction.
With VendLink up and running, route drivers only had to visit their vending machines when they needed restocking or mechanical service. Yes, we accurately predicted machine failures without Machine Learning technology. Furthermore, we learned customer preferences to deliver more of the products that people wanted which made each vending machine more profitable.
