Blockchain based transformation a gartner trend insight report pdf

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Top 10 Strategic Technology Trends for 2018 Published: 3 October 2017 ID: G00327329

Analyst(s): David W. Cearley, Brian Burke, Samantha Searle, Mike J. Walker

The intelligent digital mesh is a foundation for future digital business and its ecosystems. To create competitive advantage, enterprise architecture and technology innovation leaders must evaluate these top trends to identify opportunities that their organizations can exploit.

Key Findings ■ Artificial intelligence (AI) delivers value to every industry, enabling new business models. It does

so by supporting key initiatives such as customer engagement, digital production, smart cities, self-driving cars, risk management, computer vision and speech recognition.

■ As people, places, processes and "things" become increasingly digitalized, they will be represented by digital twins. This will provide fertile ground for new event-driven business processes and digitally enabled business models and ecosystems.

■ The way we interact with technology will undergo a radical transformation over the next five to 10 years. Conversational platforms, augmented reality, virtual reality and mixed reality will provide more natural and immersive interactions with the digital world.

■ A digital business is event-centric, which means it must be continuously sensing and adapting. The same applies to the security and risk infrastructure that supports it, which must focus on deceiving potential intruders and predicting security events.

Recommendations Enterprise architecture (EA) and technology innovation leaders using EA to master emerging and strategic trends must:

■ Devise new business scenarios using AI as the enabler for new business designs. Do so by engaging, educating and ideating with senior business leaders about their strategically relevant priorities.

■ Create a more natural and immersive user experience by deploying, where effective, conversational platforms and virtual, augmented and mixed reality.

■ Support Internet of Things (IoT) initiatives by developing and prioritizing targeted, high-value business cases to build digital twins and exploit cloud and edge computing synergistically.

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

■ Adopt a strategic approach for security and risk that continuously adapts based on risk and trust. Do so by communicating requirements to developers, achieving a DevSecOps environment.

Table of Contents

Analysis..................................................................................................................................................3

Trend No. 1: AI Foundation...............................................................................................................4

Today's AI Is Narrow AI...............................................................................................................5

Trend No. 2: Intelligent Apps and Analytics....................................................................................... 6

Augmented Analytics Will Enable Users to Spend More Time Acting on Insights.........................8

Trend No. 3: Intelligent Things...........................................................................................................9

Swarms of Intelligent Things Will Work Together....................................................................... 11

Trend No. 4: Digital Twins............................................................................................................... 12

Digital Twins Will Be Linked to Other Digital Entities...................................................................14

Trend No. 5: Cloud to the Edge...................................................................................................... 15

Edge Computing Brings Distributed Computing Into the Cloud Style........................................ 16

Trend No. 6: Conversational Platforms............................................................................................17

Integration With Third-Party Services Will Further Increase Usefulness...................................... 18

Trend No. 7: Immersive Experience.................................................................................................20

VR and AR Can Help Increase Productivity............................................................................... 21

Trend No. 8: Blockchain................................................................................................................. 23

Blockchain Offers Significant Potential Long-Term Benefits Despite Its Challenges....................24

Trend No. 9: Event-Driven Model.................................................................................................... 26

Events Will Become More Important in the Intelligent Digital Mesh............................................ 26

Trend No. 10: Continuous Adaptive Risk and Trust......................................................................... 27

Barriers Must Come Down Between Security and Application Teams....................................... 28

Gartner Recommended Reading.......................................................................................................... 30

List of Figures

Figure 1. Top 10 Strategic Technology Trends for 2018...........................................................................4

Figure 2. Narrow AI's Place in the Long History of AI.............................................................................. 5

Figure 3. Augmented Analytics for Citizen and Professional Data Scientists............................................ 8

Figure 4. Intelligent Things Span Many Sectors.....................................................................................10

Figure 5. Digital Twins Are Digital Representations of Real-World Objects............................................. 13

Page 2 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Figure 6. Digital-Twin Models Will Expand to More Than Just Things.................................................... 15

Figure 7. Cloud and Edge Computing Are Complementary Concepts...................................................17

Figure 8. Conversational Platforms Include New User Experience Design Elements..............................19

Figure 9. The Future of the User Experience......................................................................................... 22

Figure 10. Key Elements of Blockchain.................................................................................................24

Figure 11. Event-Driven and Request-Driven Application Design Models Are Complementary.............. 27

Figure 12. The DevSecOps Model........................................................................................................ 29

Analysis Digital business blurs the physical and virtual worlds in a way that transforms business designs, industries, markets and organizations. The continuing digital business evolution exploits emerging and strategic technologies to integrate the physical and digital worlds, and create entirely new business models. The future will be defined by smart devices delivering increasingly insightful digital services everywhere. We call this mesh of interconnected people, devices, content and services the intelligent digital mesh. It's enabled by digital business platforms delivering a rich intelligent set of services to support digital business. As an EA or technology innovation leader seeking to exploit the intelligent digital mesh, you must respond to the disruptive technology trends driving this future.

Our top 10 strategic technology trends include three groupings of complementary trends (see Figure 1):

■ The intelligent theme explores how AI is seeping into virtually every existing technology and creating entirely new technology categories. The exploitation of AI will be a major battleground for technology providers through 2022. Using AI for well-scoped and targeted purposes delivers more flexible, insightful and increasingly autonomous systems.

■ The digital theme focuses on blending the digital and physical worlds to create a natural and immersive, digitally enhanced experience. As the amount of data that things produce increases exponentially, compute power shifts to the edge to process stream data and send summary data to central systems. Digital trends, along with opportunities enabled by AI, are driving the next generation of digital business and the creation of digital business ecosystems.

■ The mesh theme refers to exploiting connections between an expanding set of people and businesses — as well as devices, content and services — to deliver digital business outcomes. The mesh demands new capabilities that reduce friction, provide in-depth security and respond to events across these connections.

Our top 10 list highlights strategic trends that aren't yet widely recognized but have broad industry impact and significant potential for disruption. Through 2022, technologies related to these trends will reach a level of maturity that crosses a critical tipping point. And they'll experience significant changes. Examine the business impact of our top 10 strategic technology trends, and seize the opportunities to enhance your existing products, create new ones or adopt new business models.

Gartner, Inc. | G00327329 Page 3 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Digital business will transform your industry. Prepare for the impact of digital business on your industry and your business.

Figure 1. Top 10 Strategic Technology Trends for 2018

Source: Gartner (October 2017)

Trend No. 1: AI Foundation

Interest in AI is growing, as shown by an increase of more than 500% in the number of inquiry calls

from Gartner clients about topics related to AI in the past year. 1 A 2017 Gartner survey found that

59% of organizations are still gathering information to build their AI strategies, while the rest have

already made progress in piloting or adopting AI solutions. 2 Furthermore, the market indicates

strong investment in startups selling AI technologies. 3

Creating systems that learn, adapt and potentially act autonomously will be a major battleground for technology vendors through at least 2020. The ability to use AI to enhance decision making, reinvent business models and ecosystems, and remake the customer experience will drive the payoff for digital initiatives through 2025.

The AI foundation consists of numerous technologies and techniques that have grown over many years. These include expert systems, decision trees, linear regression and neural networks. The level of capability has grown steadily. This is the result of:

■ Ever-more advanced algorithms using supervised, unsupervised and reinforcement-learning techniques

■ The availability of massive amounts of data to feed machine learning

Page 4 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

■ Hardware advances (such as servers based on graphics processing units) delivering massive compute infrastructure to process the huge amount of data and sophisticated algorithms

Advanced machine learning in the form of deep learning has further extended the problem domains that AI addresses. Examine the wide variety of AI-related techniques and exploit them as needed.

Today's AI Is Narrow AI

Today, the focus for AI is on "narrow AI" (see Figure 2). Narrow AI consists of highly scoped machine-learning solutions that target a specific task (such as understanding language or driving a vehicle in a controlled environment). The algorithms chosen are optimized for that task. All the real- world examples of AI in use or development are examples of narrow AI. Artificial general intelligence refers to the use of machine learning to handle a broad range of use cases. Such systems, were they to exist, would successfully perform any intellectual task that a human could perform and would learn dynamically, much as humans do. These systems may never exist, but interest in them continues in the popular media and among those predicting an "AI doomsday." Focus on business results enabled by applications that exploit narrow AI technologies, both leading-edge and older AI technologies. Leave general AI to the researchers and science fiction writers.

Evaluate a number of business scenarios in which AI could drive specific business value, and consider experimenting with one or two high-impact scenarios. For example, in banking, you could use AI techniques to model current real-time transactions, as well as make predictive models of transactions based on their likelihood of being fraudulent. If you're an early adopter or you're seeking to drive disruptive innovation, begin to implement predictive analytics, ensemble learning and natural-language processing. If you're a mainstream user or have more modest innovation goals, use third parties and packaged solutions with embedded AI (see "Ten Ways AI Will Appear in Your Enterprise — No One Source Can Meet All Your Business Needs").

Figure 2. Narrow AI's Place in the Long History of AI

Source: Gartner (October 2017)

Gartner, Inc. | G00327329 Page 5 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

AI techniques are evolving rapidly. You'll need to invest significantly in skills, processes and tools to successfully exploit these techniques. Investment areas include setup, integration, algorithm/ approach selection, data preparation and model creation. In addition, it can take significant effort to exploit a system's learning capabilities, evaluate the accuracy of findings, and update the algorithms and models to improve results. Effort is required from not only the data scientists creating the system, but also others who have the knowledge needed to "train" the system. You'll need:

■ Data scientists to understand data and AI algorithms, and to formulate coherent questions or problem domains to which to apply these algorithms

■ Application developers to design interfaces, services and process flows

A lack of the relevant data sciences will probably hamper AI adoption in the short term. 4 By 2020,

30% of new development projects will deliver AI through joint teams of data scientists and programmers.

Applied AI gives rise to a range of intelligent implementations. These include physical devices (such as robots, autonomous vehicles and consumer electronics), as well as apps and services (such as virtual personal assistants [VPAs] and smart advisors). These implementations will be delivered as a new class of obviously intelligent apps and things. They'll provide embedded intelligence for a wide range of mesh devices, and existing software and service solutions. The data science needed to create these systems is complex. This means that many organizations will consume applied AI mainly through packaged intelligent apps and things. Alternatively, organizations will consume them through packaged platform services or "models as a service" that they can build into custom applications.

Related Research:

■ "Develop Your Artificial Intelligence Strategy Expecting These Three Trends to Shape Its Future"

■ "AI on the Edge: Fusing Artificial Intelligence and IoT Will Catalyze New Digital Value Creation"

■ "Market Trends: How AI and Affective Computing Deliver More Personalized Interactions With Devices"

■ "Applying Artificial Intelligence to Drive Business Transformation: A Gartner Trend Insight Report"

■ "Innovation Insight for Artificial Intelligence of Things — Machine Learning in the IoT Era"

■ "Where You Should Use Artificial Intelligence — and Why"

■ "Questions to Ask Vendors That Say They Have 'Artificial Intelligence'"

Trend No. 2: Intelligent Apps and Analytics

Organizations are applying AI techniques to create new app categories (such as virtual customer assistants [VCAs]) and improve traditional applications (such as worker performance analysis, sales and marketing, and security). Intelligent apps have the potential to transform the nature of work and

Page 6 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

the structure of the workplace. When building or buying an AI-powered app, consider where its AI impact will be. It's useful to focus on three target domains when exploring how and where to exploit AI:

■ Analytics: AI can be used to create more predictive and prescriptive analytics that can then be presented to users for further evaluation, or plugged into a process to drive autonomous action. AI is also being used for augmented analytics.

■ Process: AI can drive more intelligent actions by an application. For example, you can use AI for intelligent invoice matching or analysis of email documents to improve service flow. In the future, this can be extended further to identify patterns of work, from which process models can be built and executed.

■ User Experience: Natural-language processing used to create VPAs is one application of AI to the user experience. Further examples include facial recognition and other AI applications for understanding user emotions, context or intent, and predicting user needs.

During the next few years, virtually every app, application and service will incorporate some level of AI. Some of these apps will be obvious intelligent apps that couldn't exist without AI and machine learning. Others will be unobtrusive users of AI that provide intelligence behind the scenes.

There is an AI "land grab" from the large vendors making "big bets" and from startups seeking to gain an edge. They all aim to support or replace manual human-based activities with intelligent automation. Vendors such as Salesforce, SAP, Oracle and Microsoft are incorporating more advanced AI functions in their offerings. These vendors are exploiting AI to varying degrees, but they're all focusing on their traditional strongholds. For example, the main enterprise software vendors are emphasizing sales, service, marketing and ERP as particularly valuable areas for applying AI techniques. Microsoft is focusing on Office 365 and a strong developer ecosystem. Challenge your packaged software and service providers to outline how they'll be using AI to add business value in new versions. Explore how much of the new value will come from bleeding-edge, rather than older, AI technologies. Examine how they use AI to deliver advanced analytics, intelligent processes and new user experiences.

VPAs such as Google Now, Microsoft's Cortana and Apple's Siri are becoming smarter and are a rapidly maturing type of intelligent app. Some chatbots, such as Facebook Messenger, can be powered by AI (for example, Wit.ai) to deliver an intelligent app. These intelligent apps feed into the conversational platform trend to create a new intelligent intermediary layer between people and systems. If you're an early adopter or you're seeking to drive disruptive innovation, begin to implement targeted VCAs and VPAs where a high-value target persona (for example, a doctor, marketing leader or high-profit customer) could achieve significant benefit. If you're a mainstream user or have more modest innovation goals, consider more simple rule-based chatbots. Exploit prepackaged assistants or simple mobile assistants based on the VPA capabilities embedded in smartphones.

Intelligent apps can create a new intelligent intermediary layer between people and systems. They have the potential to transform the nature of work and the structure of the workplace, as seen with VCAs and enterprise advisors and assistants. These models free people to build on and extend the

Gartner, Inc. | G00327329 Page 7 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

capabilities of the assistant. For example, in healthcare, advanced advisors and other AI-assisted capabilities have the potential to enhance doctors' understanding and their ability to deliver more personalized treatments. Explore intelligent apps as a way of augmenting human activity, and not simply as a way of replacing people.

Augmented Analytics Will Enable Users to Spend More Time Acting on Insights

Augmented analytics is a particularly strategic, next-generation data and analytics paradigm in which AI is having an impact (see Figure 3). It uses machine learning to automate data preparation, insight discovery and insight sharing for a broad range of business users, operational workers and citizen data scientists. Augmented analytics will enable expert data scientists to focus on specialized problems and on embedding enterprise-grade models into applications. Users will spend less time exploring data and more time acting on the most relevant insights. They will do so with less bias than in manual approaches.

Figure 3. Augmented Analytics for Citizen and Professional Data Scientists

NLG = natural-language generation; NLP = natural-language processing; NLQ = natural-language query

Source: Gartner (October 2017)

Enterprises will need to develop a strategy to address the impact of augmented analytics on currently supported data and analytics capabilities, roles, responsibilities and skills. They'll also need to increase their investments in data literacy. Both small startups and large vendors now offer augmented analytics capabilities that could disrupt vendors of business intelligence and analytics, data science, data integration, and embedded analytic applications. Data and analytics leaders

Page 8 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

must review their investments. By 2020, augmented analytics will be the dominant driver for data analysis systems. And by 2020, automation of data science tasks will enable citizen data scientists to produce a higher volume of advanced analysis than specialized data scientists.

Intelligent apps constitute a long-term trend that will evolve and expand the use of AI in apps and services through 2037. Establish a process to continually evaluate where your organization can apply AI today and over time. Use persona-based analysis to determine the most appropriate opportunities. Compare the roadmaps for AI exploitation across your packaged app and service provider portfolio. Proceed with caution if your organization is developing applications — the underlying AI elements for creating intelligent apps aren't ready for most application development projects at scale. Ensure such projects have a very high potential business value. The competitive gaps and missed opportunity costs for laggards could be significant.

Related Research:

■ "Market Guide for Virtual Customer Assistants"

■ "Competitive Landscape: Virtual Personal Assistants, 2016"

■ "Augmented Analytics Is the Future of Data and Analytics"

■ "Hype Cycle for Analytics and Business Intelligence, 2017"

■ "How Enterprise Software Providers Should (and Should Not) Exploit the AI Disruption"

Trend No. 3: Intelligent Things

Intelligent things are physical things that go beyond the execution of rigid programming models and exploit AI to deliver advanced behaviors that interact more naturally with their surroundings and with people. AI is driving advances for new intelligent things, such as autonomous vehicles, robots and drones, and delivering enhanced capability to many existing things, such as IoT-connected consumer and industrial systems (see Figure 4).

Gartner, Inc. | G00327329 Page 9 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Figure 4. Intelligent Things Span Many Sectors

Source: Gartner (October 2017)

Intelligent things are either semiautonomous or fully autonomous. The word "autonomous," when used to describe intelligent things, is subject to interpretation. When Gartner uses this term to describe intelligent things, we don't mean that these intelligent things have AI-style freedom from external human control or influence. Rather, we mean that these intelligent things can operate unsupervised for a defined period to complete a task. Intelligent things may have various levels of autonomy, as shown by the following examples:

■ Self-directing vacuum cleaners that have limited autonomy and smartness

■ Drones that can autonomously dodge obstacles 5

■ Unmanned aerial vehicles that can fly into buildings through windows and doors

Autonomous drones and robots will undergo significant technical evolution powered by new machine-learning models and algorithms. They will be used mainly in narrowly defined scenarios and controlled environments. Advances in one domain — such as more sophisticated algorithms that enable a robot to learn from its environment — will often have an application in another domain.

The use of autonomous vehicles in controlled settings (for example, farming, mining and warehousing) is a growing area of interest for intelligent things. In industrial settings, vehicles can be fully autonomous. By 2022, it's likely that autonomous vehicles will be used on roadways in limited, well-defined, geofenced and controlled areas. But general use of autonomous cars will probably require a person in the driver's seat in case the technology should fail — several U.S. states have

Page 10 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

passed laws stipulating this. In the near term, high-technology companies and traditional automotive manufacturers (such as Ford, Uber, Alphabet's Google, Volkswagen, Mercedes-Benz, Tesla, Nissan, BMW and Honda) will all be testing autonomous vehicles. Through at least 2022, we expect that semiautonomous scenarios requiring a driver will dominate. During this time, manufacturers will test the technology more rigorously, and the nontechnology issues will be addressed, such as regulations, legal issues and cultural acceptance.

AI will be embedded more often into everyday things, such as appliances, speakers and hospital equipment. This phenomenon is closely aligned with the emergence of conversational platforms, the expansion of the IoT and the trend toward digital twins. Amazon Echo is an example of an intelligent thing. It's a simple speaker connected wirelessly to an assistant, powered by machine learning. As conversational interfaces are delivered through other devices with a speaker or text input option, all these objects will become intelligent things.

Other markets have similar potential for embedded intelligence. For example, today's digital stethoscope can record and store heartbeat and respiratory sounds. Collecting a massive database of such data, relating the data to diagnostic and treatment information, and building an AI-powered doctor assistance app would enable doctors to receive diagnostic support in real time. However, in more advanced scenarios, significant issues such as liability, patient privacy and regulatory constraints must be considered. We expect that these nontechnical issues, and the complexity of creating highly specialized assistants, will slow embedded intelligence in industrial IoT and other business scenarios. Organizations that can address these barriers have the potential for significant competitive advantage.

Swarms of Intelligent Things Will Work Together

As intelligent things proliferate, we expect a shift from stand-alone intelligent things to a swarm of collaborative intelligent things. In this model, multiple devices will work together, either independently of people or with human input. For example, if a drone examined a large field and found that it was ready for harvesting, it could dispatch an "autonomous harvester." In the delivery market, the most effective solution may be to use an autonomous vehicle to move packages to the target area. Robots and drones on board the vehicle could then effect final delivery of the package. The military is leading the way in this area and is studying the use of drone swarms to attack or

defend military targets. 6 Other examples include:

■ Intel's use of a drone swarm for the U.S. Super Bowl halftime show in 2017 7

■ A plan for Dubai to use autonomous police vehicles that can deploy their own drones for

surveillance 8

■ Cooperative merge scenarios by Honda and other car manufacturers, in which vehicles

communicate with one another to optimize traffic flows 9

Related Research:

■ "Use Scenarios to Plan for Autonomous Vehicle Adoption"

Gartner, Inc. | G00327329 Page 11 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

■ "Supply Chain Brief: Favorable Regulations Will Accelerate Global Adoption of Autonomous Trucking"

■ "Predicts 2017: Drones"

■ "Hype Cycle for Drones and Mobile Robots, 2017"

■ "Market Trends: Personal Assistant Robots for the Home"

■ "Swarms Will Help CIOs Scale Up Management for Digital Business"

Trend No. 4: Digital Twins

A digital twin is a digital representation of a real-world entity or system (see Figure 5). The implementation of a digital twin is an encapsulated software object or model that mirrors a unique physical object (see Note 1). Data from multiple digital twins can be aggregated for a composite view across a number of real-world entities. The notion of a digital representation of real-world entities or systems is not new. You can argue that this was a central notion in the IT industry with the creation of computer-aided design representations of physical assets or profiles of individual customers. The difference in the latest iteration of digital twins is:

■ The robustness of the models

■ Digital twins' link to the real world, potentially in real time

■ The application of advanced big data analytics and AI

■ The ability to interact with them and evaluate "what if" scenarios

Digital twins in the context of IoT projects are leading the interest in digital twins today. 10

Well- designed digital twins of assets could significantly improve enterprise decision making. They are linked to their real-world counterparts and are used to understand the state of the thing or system, respond to changes, improve operations, and add value (see Figure 5).

Page 12 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Figure 5. Digital Twins Are Digital Representations of Real-World Objects

CAD = computer-aided design; FEA = finite element analysis; ML = machine learning

Source: Gartner (October 2017)

By 2020, we estimate there will be more than 20 billion connected sensors and endpoints, 11

and digital twins will exist for potentially billions of things. Benefits will include asset optimization, competitive differentiation and improved user experience in nearly all industries. As OEMs continue to work on connected products, they'll need to do more than just provide digital twins of their assets based on the essential elements described in Note 1. Rather, OEMs will need to think about their customers' evolving use cases and business models. Only by doing this can OEMs ensure that their hardware and software products remain competitive.

Organizations will implement digital twins simply at first. They will evolve them over time, improving their ability to collect and visualize the right data, apply the right analytics and rules, and respond effectively to business objectives. Through 2027, digital-twin use will expand beyond product engineers and data scientists. Operations managers will use them for assets where the cost-benefit analysis of risks in operations makes the case for digital twins compelling. We also expect that digital-twin models will proliferate, with suppliers increasingly providing customers with these models as an integral part of their offering.

Digital twins can enhance data insights and improve decision making, and will eventually help in the development of new business models. Their use will bring numerous benefits in different time frames, including:

■ Short term: Digital twins help in asset monitoring, optimization and improving the user experience, which is vital in nearly all industries. The shift from preventive to predictive (condition-based) maintenance is a particularly high-value use of digital twins. Customer

Gartner, Inc. | G00327329 Page 13 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

benefits include reducing maintenance-driven downtime and lowering operating and maintenance costs.

■ Midterm: Organizations will use digital twins to operate factories and increase operational efficiency. They will use them to plan for equipment service and to predict equipment failure, enabling them to repair equipment to prevent its failure. Organizations will also use digital twins to enhance product development. They will do this by using them to simulate the behavior of new products based on the digital-twin insight from previous products, taking into consideration their cost, environment and performance. Digital twins offer the possibility of business models centered on guaranteed outcomes, such as specific asset performance guarantees.

■ Long term: Digital twins will aid innovation by providing insights into how products and services are used and how they can be improved. New business models may center on proactive advice. For example, automotive engineers could use digital twins, in conjunction with an analytics tool, to analyze how a specific car is driven to suggest new features to reduce accidents. Engineers might also suggest new products to serve the machine as a customer, where the machine and its digital twin have a budget for approved services. Other models may center on potential new marketplaces for digital twins, interfaces and suitably sanitized datasets from digital twins.

Digital Twins Will Be Linked to Other Digital Entities

Digital twins consolidate massive amounts of information on individual assets and groups of assets, often providing control of those assets. As the digital-twin trend evolves, twins will communicate with one another to create "digital factory" models of multiple linked digital twins. Digital twins of assets will be linked to other digital entities for people (digital personas), processes (law enforcement) and spaces (digital cities). Understanding the links across these digital entities, isolating elements where needed and tracking interactions will be vital to support a secure digital environment.

Although much attention is on digital twins of assets as part of an expanding IoT model, more sophisticated digital models of the real world have a much larger impact. Digital twins are built on the concept that virtual asset models coexist and are connected to real assets — they are twins. However, this concept isn't limited to assets (or things). Digital analogs of real-world elements are growing along many dimensions. Like digital twins, these other digital entities often grow from metadata structures and models of things in the real world that are disconnected from it, or are, at most, only loosely connected to it. Over time, these digital representations/models will be connected more tightly to their real-world counterparts. They will be infused with more sophisticated AI-based models, just as we are seeing with digital twins for assets. The following will be used for advanced simulation, operations and analysis (see Figure 6):

■ Future models of humans that could include rich biometric and medical data

■ Business operating system models defining the details of business processes and ecosystem interactions

■ Sophisticated models of buildings, cities and other places

Page 14 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Figure 6. Digital-Twin Models Will Expand to More Than Just Things

Source: Gartner (October 2017)

Related Research:

■ "Innovation Insight for Digital Twins — Driving Better IoT-Fueled Decisions"

■ "Hype Cycle for the Internet of Things, 2017"

■ "Predicts 2017: IT and OT Convergence Will Create New Challenges and Opportunities"

■ "Digital Twins Will Impact Economic and Business Models"

■ "Create a Digital Twin of Your Organization to Optimize Your Digital Business Transformation Program"

■ "Digital Connectivism Tenet 1: We All Have a Digital Identity"

Trend No. 5: Cloud to the Edge

Edge computing describes a computing topology in which information processing and content collection and delivery are placed closer to the sources and sinks of this information. Edge computing draws from the concepts of mesh networking and distributed processing. It tries to keep the traffic and processing local, with the goal being to reduce traffic and latency. As such, the notion of edge content delivery has existed for many years. The "where to process the data" pendulum has swung between highly centralized approaches (such as a mainframe or a centralized cloud service) and more decentralized approaches (such as PCs and mobile devices). Connectivity and latency

Gartner, Inc. | G00327329 Page 15 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

challenges, bandwidth constraints and greater functionality embedded at the edge favor distributed deployment models. The advantages of processing power and low costs of operating at hyperscale, coupled with the complexity of managing and coordinating thousands of geographically separated endpoints, favor the centralized model.

Much of the current focus on edge computing comes from the need for IoT systems to deliver disconnected or distributed capabilities into the embedded IoT world. Widespread application of the topology and explicit application and networking architectures aren't yet common. Systems and networking management platforms will need to be stretched to include edge locations and edge- function-specific technologies. These include data thinning, data compression and protection, and local analytics. Edge computing solves many pressing issues, such as high WAN costs and unacceptable latency. The edge computing topology will enable the specifics of digital business and IT solutions uniquely well in the near future.

Begin using edge design patterns in your mid- to longer-term infrastructure architectures. Immediate actions might include simple trials using colocation and edge-specific networking capabilities. You could also simply place remote-location or branch-office compute functions in a standardized enclosure (for example, "data center in a box"). Some applications, such as client- facing web properties and branch-office solutions, will be simpler to integrate and deploy. Data thinning and cloud interconnection will take more planning and experimentation to get right.

Edge Computing Brings Distributed Computing Into the Cloud Style

Most view cloud and edge computing as competing approaches. They view public cloud deployments as enjoying the economies of hyperscale, centralized data centers, with edge computing mandating processing to be pushed to the edge. But this is a misunderstanding of the two concepts. Cloud computing is a style of computing in which elastically scalable technology capabilities are delivered as a service using internet technologies. Cloud computing doesn't mandate centralization. Edge computing brings the distributed computing aspect into the cloud style. Consider cloud and edge computing as complementary rather than competing concepts (see Figure 7). You can use:

■ Cloud computing as a style of computing to create a service-oriented model and a centralized control and coordination structure

■ Edge computing as a delivery model, allowing for disconnected or distributed process execution of aspects of the cloud service

Page 16 of 34 Gartner, Inc. | G00327329

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

Figure 7. Cloud and Edge Computing Are Complementary Concepts

Source: Gartner (October 2017)

Some cloud implementations already use an approach that distributes functionality to the edge (for example, Microsoft Office 365 and AWS Greengrass). We expect this approach will be used more frequently as cloud vendors push further into the IoT market, and IoT solution vendors adopt the cloud style as a way to manage their solutions more effectively. Although the IoT is a strong driver for a cloud-to-the-edge approach, the trend will also benefit mobile and desktop environments. More solutions similar to Office 365 are likely to appear.

Related Research:

■ "Cool Vendors in IoT Edge Computing, 2017"

■ "Expand Your Artificial Intelligence Vision From the Cloud to the Edge"

■ "A Guidance Framework for Architecting the Internet of Things Edge"

■ "Explore the Roles of IoT Gateways in Five Edge Use Cases"

■ "The Edge Manifesto: Digital Business, Rich Media, Latency Sensitivity and the Use of Distributed Data Centers"

■ "Market Guide for Edge Computing Solutions for Industrial IoT"

Trend No. 6: Conversational Platforms

Conversational platforms will drive the next big paradigm shift in how humans interact with the digital world. They will shift the model from technology-literate people to people-literate technology.

Gartner, Inc. | G00327329 Page 17 of 34

This research note is restricted to the personal use of cletavec1@udayton.edu.

This research note is restricted to the personal use of cletavec1@udayton.edu.

The burden of translating intent will move from the user to the computer. The system takes a question or command from the user in natural language. It responds by executing a function, presenting content or asking for additional input.

A conversational platform provides a high-level design model and execution engine in which user and machine interactions occur. As the term "conversational" implies, these interfaces are implemented mainly in the user's spoken or written natural language. In time, other input/output mechanisms will be added to exploit sight, taste, smell and touch for multichannel interaction. The use of expanded sensory channels will support advanced capabilities, such as emotion detection through facial expression analysis and human health status through olfactory analysis. But exploitation of these other sensory channels will be isolated and limited for the next three to five years.

Over the next few years, conversational interfaces based on natural-language interfaces will become the main design goal for user interaction. Gartner predicts that, by 2019, 20% of users'

interactions with smartphones will be through VPAs. 12

A Gartner survey found that a quarter of

smartphone users had used their VPA in the past month, most on a daily or weekly basis. 13

Conversational platforms are most recognizably implemented in:

■ VPAs, such as Amazon Alexa, Apple's Siri, Google Assistant and Microsoft's Cortana

■ VCAs, such as IPsoft's Amelia, Watson Virtual Agent, and VCAs from [24]7, Artificial Solutions, Interactions, Next IT and Nuance

■ Chatbot frameworks, such as Amazon Lex, API.AI from Google, IBM Watson Conversation and Microsoft Bot Framework

Interactions in conversational platforms are typically informal and bidirectional. The interaction may be a simple request or question (such as "What's the weather forecast?" or "What time is it?") with a simple result or answer. Alternatively, it may be a structured interaction (such as that required to book a restaurant table or hotel room). As the technology matures, extremely complex requests will be possible, resulting in highly complex results. For example, the conversational platform may be able to collect oral testimony from crime witnesses, resulting in the creation of a suspect's image.

Integration With Third-Party Services Will Further Increase Usefulness

Conversational platforms have reached a tipping point: the usefulness of the systems has exceeded the friction of using them. But they still fall short. Friction is created when users need to know which domains the UI understands and what its capabilities are within those domains. The challenge that conversational platforms face is that users must communicate in a very structured way. This is often a frustrating experience. Rather than enabling a robust two-way conversation between the person and the computer, most conversational platforms are mainly one-directional query or control systems that produce a very simple response. Over time, more conversational platforms will integrate with growing ecosystems of third-party services that will exponentially drive the usefulness of these systems. A primary differentiator among conversational platforms will be the robustness of their conversational models and the API and event models used to access, invoke and orchestrate third-party services to deliver complex outcomes.