Big Data Enabling In Industrial Internet of Things

Info: 18184 words (73 pages) Dissertation
Published: 10th Dec 2019

Reference this

Tagged: Internet of Things

Share this: Facebook Twitter Reddit LinkedIn WhatsApp

Faculty of Business and Management

CONTENTS

INTRODUCTION

FUNDAMENTALS 7-11

INTERNET OF THINGS AS ORIGIN 1-2

INDUSTRIAL INTERNET OF THINGS BASICS 2-3

INTRODUCTION

The rapid development of the Internet of Things (IoT) has influenced many spheres of the life and now is gaining popularity in a number of industries, as manufacturing, gas and oil, agriculture and more. Industrial IoT (IIoT) as a new trend [1] is attracting a lot of attention from industrial and scientific societies, as it is promising the optimization of asset utilization (e.g. remote management for reducing failures or downtime) and creation of new revenue streams via automation of operational processes. Originally, General Electric [2] described Industrial Internet as a connection of devices over the Internet conducting physics-based analytics, predictive modeling and in general the automation of the manufacturing operations by means of advanced sensors, controls, special software, and support operations available anytime and from any place. However, with the invention of specific solutions and further attempts to develop IIoT idea towards the beginning of the fourth industrial revolution [3], the definition has changed a lot. Reviews from consulting companies and machine-to-machine innovators consider it from different perspectives and, thus lead to misunderstanding of the concept. The following is intended: in some cases, the IIoT means the data that might be obtained as a result of manufacturing processes, integration of the sources and analytics that stands for starting out from data analytics [4]. Others describe a set of sensors installed at the factory facilities, recording the data and transferring it to the information systems that provide workers with necessary information about the process state [5]. This viewpoint mostly describes the work of distributed embedded systems, but let the term “internet” in its initial meaning be missing from the abbreviation. Consequently, this brings up the question of understanding the idea of IIoT, as real application suggested by appearing digital technological startups [6], often differs from the theoretical model of a broader IoT concept.

A high relation between IIoT and Big Data is another important issue. “Sensors embedded in end-points are not much help if the data they generate can’t be collected and transmitted for analysis” – notices the analyst from Moor Insights & Strategy [7]. In point of fact, the already well-established idea of consumer Internet of Things is aimed at improving human life, making it healthy, safe and enjoyable, while the Industrial IoT is associated with delivery of a service and implies utilization by a professional or an enterprise with the original goal to increase productivity and enable maintenance services, energy and design optimization applications, thus it is data-driven.

Research Questions

The research that will be conducted in this paper is intended to clarify the term of IoT and IIoT in general. Additionally, the connection of these terms to Big Data should be addressed, to reveal the applications of Big Data in this concern.

Is Big Data indivisible bundling with IoT or is it just used as a supplement that can improve IIoT? After inducing a general understanding of Internet of Things, Industrial IoT and Big Data in the same context, this paper will show up issues that should be analyzed in detail and be dealt with utter care when IIoT is planned to be implemented in the industry.

The master thesis is devoted to the research of Industrial Internet of Things and especially the usage of information obtained from the connected and sharing data physical objects, including older equipment. Taking into account the industry, the data participating in the IIoT communications often refers to Big Data term due to large volumes, different type and nature and specific requirements for the processing. Thus, the main aim of the present paper is to analyze the enabling of Big Data implication in the Industrial Internet of Things, considering theoretical researches and existing solutions suggested by IoT technologies’ providers.

The main hypothesis implies that there is a strong correlation between Industrial Internet of Things and Big Data, but the ambivalent definition of Industrial Internet and market-speak are causing the appearance of solutions having no connection with both mentioned concepts. The paper will test this hypothesis and

The research questions are the following:

What is the concrete definition of Industrial Internet of Things?
1. What are the special aspects of IIoT that make it different from the broader IoT term?
2. How is IIoT represented by the modern digital technology providers?
How is Big Data linked to IIoT and what are its possible implications?
1. How is the data in IIoT collected/preprocessed/stored/processed?
2. In which context can this data be referred to Big Data?
3. How is Big Data currently applied in IIoT?
Which misunderstandings occur in the industry while applying Big Data IIoT solutions?
1. How is the current state of the IIoT solutions market?
2. Do the found cases meet the requirements to be called Big Data IIoT solutions? Why/Why not?
3. What recommendations can be made for the companies interested in implementing IIoT?

Methodological Approach

Method 1 is literature review conducted with use of scientific papers, companies’ reports and documents from conferences. This method will help to create the detailed definition of IIoT

Method 2 is the content analysis of the modern IoT providers’ solutions aimed at the description of real IIoT application

Method 3 is the analysis of case studies and conducting expert interviews with IIoT solutions providers for assessment of real implication of Big Data and IoT in their products

Method 4 is the creation of the theoretical model summarizing the task description statements

STRUCTURE OF THE WORK

The first part is devoted to explore the concept of IIoT and create the detailed definition of the term, taking into account the viewpoint of different IoT providers. One of the biggest issues here is the research the coexistence of IIoT and Big Data. Taking existing case studies as a basis, the challenge is to set the benefits of the Big Data analytics for companies of various industrial sectors and introduce the special aspects that can be used to identify Big Data in the industry – but not just large amounts of data that are collected.

The second part will concretize the previously elaborated IIot aspects, for example to look into the data collecting process and to determine which data should be used in what way. The research will cover issues as data gathering, storing, and processing and the whole IIoT architecture [8], including physical objects, middleware, software applications and human operations. Defining sources, ways of extracting data, its representation and transmitting techniques, the work will be focused on the description of possible use cases of the obtained data in Big Data analytics.

Another task is to conduct marketing analysis aimed at the discovery of IIoT solutions that exist on the market.

Considering the ideas concluded on the previous steps, analysis if companies really provide IIoT and how their solutions deal with Big Data will be conducted. Web-sites from organizations and startups that identify themselves as IIoT providers will be used for this content analysis. To further validate and to additionally refine the outcomes of the analysis interviews with above mentioned IIoT providers will be held, questioning them about their vision of IIoT and Big Data in their products.

In the last part of this paper a theoretical model should be described based on the parts that were elaborated previously, which shows up possible ways of implementing IIoT in combination with Big Data in the fields of industry. This model will help to find answers to for example following questions: what opportunities companies can get from its application, what are the main technical requirements and limits. The model will consider all special characteristics of real Big Data IIoT solutions and conclude the approach that can be used for evaluation of Big Data enabling in Industrial Internet solutions. It will include aspects of devices usage, the architecture of gathering, storage and transferring data and the approaches for conducting the analysis.

This paper may be useful to those who are interested in the current trends in Information Technologies sphere with both technical or business backgrounds, as the conducted research will cover not only engineering components of Industrial Internet of Things but also marketing issues represented by the review of the existing solutions. Although the paper will provide a reader with all necessary explanations and definitions, the audience is desired to possess some background knowledge about IoT and Big Data (at least terms used in these spheres) and business informatics in general. The model created on the last step of the work might be useful for the entrepreneurship that considers implementing IIoT in their business as an approach for choosing the provider and a list of features that the attention should be paid to.

The introduction has to be short (1-3 pages) and concrete. It must be clear to what the work is dedicated to, what goals are present and what are the methods of solving them.
Goals of the research determine its direction in which author develops the subject of the course project work. Thus the title of the work is usually the main as the main objective of the research

The first section, usually has a theorectical purpose, in which the author makes a critical analysis of the given material and draws their conclusions that would later be used in resolution of the given problems.

FUNDAMENTALS 7-11

The main research question for this part of the thesis is ‘What is the concrete definition of Industrial Internet of Things?’ Based on the literature review, it is planned to create the detailed definition of IIoT and list the main characteristics of its application. The content analysis will provide with the information on the current state of IIoT and the supplement for its defining based on the real examples via analyzing the approaches of different companies to describing and implementing this idea. To answer this question, firstly the broader term of Internet of Things will be analyzed as a basis of Industrial Internet. Clarifying all necessary features (e.g. the use of sensors), industries of appliance and opportunities for the companies the preconditions for IIoT implementation will be stated. It is expected that the section will cover some issues that imply the foundation for accomplishing the main objective of the thesis.

INTERNET OF THINGS AS ORIGIN 1-2

The Internet of Things is defined by ITU-T Y.2060 [9] as a global network infrastructure for the information society that implies self-configurable options and enables advanced services with use of interconnected physical and virtual things. However, it is needed to look at the definition in more details [10] [11]. The core of the IoT are things (T) that can be represented by devices, machines, buildings, etc. Firstly, they are not networked, have no association with either virtual or physical things and are driven by means of mechanical power or some chemical processes for the performing of their original function. As far as they are equipped with electronic features and software they can become operated by their own algorithms or some external systems. Embedding sensors allow the thing to have a feeling to its environment, so that it can use the conditions of its environment in the decision-making processes and the communication with other components of IoT system. Thus, the things must have the following characteristics:

Things can gather the data from their environment – usually via sensors, for instance, measure the conditions in their environment as temperature, pressure, humidity, etc.
Things are capable to perform calculations with the use of software integrated into the physical structure of things, their electronics.
Things are interconnected, so that they data and commands can be transmitted over the whole network.

The last point states for the word Internet (I) in the abbreviation. To have a possibility to bind all the things into a network the different ways of connecting can be used. These are already well-known technologies as Wireless Local Area Network (WLAN), 4G, Global System for Mobile Communications (GSM), Bluetooth as well as Near Field Communication (NFC) [12] that allows high-frequency short-range communication between devices and the lightweight protocol for messaging between sensors and small mobile devices Message Queue Telemetry Transport (MQTT) [13].

Nowadays, the Internet of Things is mostly represented by smart home and car, smart grids and initiatives such as the Cisco Planetary Skin [14], the online platform for the global ecological monitoring.

Smart home concept includes different solutions for intelligent security services and services to optimize the use of household resources, for example Microsoft [15]. Smart cars

• Class fleet management services for individual carriers (some analogue of Uber for freight transport)

• UBI-insurance services

• Maintenance Services for the actual state

3. Commercial and financial services:

• Solutions for the automatic transmission and analysis of data with POS-terminals, including virtual

• Inventory Management households as a service.

Finnaly, it is applied in the industrial segment, changing automatic process control system principles to IoT.

Сегодня Интернет вещей живет и здравствует, чему в немалой степени способствуют такие инициативы как Cisco Planetary Skin, Smart Grid и появление “умных” автомобилей.

1. «Умный дом», включая:

Решения для создания интеллектуальных сервисов безопасности
Решения для создания интеллектуальных сервисов оптимизации использования ресурсов домохозяйствами

2. «Умный транспорт», включая:

Сервисы класса fleet management для индивидуальных перевозчиков (некий аналог Uber для грузового транспорта)
Сервисы UBI-страхования
Сервисы технического обслуживания по фактическому состоянию

3. Торговля и финансовые услуги:

Решения для автоматической передачи и анализа данных с POS-терминалов, включая виртуальные
Управление запасами домохозяйств как сервис.

4. Промышленный сегмент – перевод АСУТП на принципы IoT.

Consumer IoT Systems. These systems connect things that consumers would typically buy directly at a store for their personal use, such as electronics, fitness devices, home automation/security devices, and leisure/entertainment/lifestyle things. Their purpose is to improve lives by making them healthier, safer, or more enjoyable. These applications are what people generally think of when they hear “the IoT.”

Industrial IoT Systems. Here, it’s the connection of things that basically are nonconsumer—the ones typically purchased by a professional or a company in order to use them in the delivery of a service. These include things like industrial machinery; transportation equipment (cars, trains, and planes); health care equipment; and megasystems like smart buildings, smart cities, and smart utility grids. Their purpose is to increase productivity, allow manufacturers to differentiate through the offering of attached services, and reduce environmental impact. Industrial IoT systems enable applications such as predictive maintenance services, energy optimization, and design optimization.

Mechanical hardware that provides structural integrity and keeps everything in place (shown in the green box below).

Electrical hardware, for example microprocessors or microcontrollers, current source, data storage and communications connections (shown in light blue box).

Electromechanical hardware such as sensors, actuators and various output devices that convert electrical energy into mechanical and vice versa (shown in yellow box).

A microcontroller with operating system running an embedded software (shown in the purple box).

The development of networked things is challenging: designers from different areas, who usually use different tools and each contribute a small part to the solution of the main problem, have to work closely together and closely coordinate.

Recently, I’ve met so many people asking about the difference between the Internet of Things and distributed embedded systems that I’ve approached our new CTO and asked him to explain it. Here’s what he replied:

“IoT is primarily a marketing, not an engineering term. When we refer to it in the engineering terms, we normally mean an embedded microprocessor controlled system connected directly or indirectly to the web (e.g., web cameras, smart thermostats, health monitors, etc.).

On the contrary, a typical distributed embedded system such as a commercial building’s lighting or heating system or an enterprise control system is never on the public Internet for obvious IT security reasons. As such, IoT refers to a highly secure and well protected embedded system that’s fully controlled by microprocessors. That’s one of the notions.

Further, IoT is a conceptual framework or an architecture that considers how components (e.g. devices) will communicate with each other (enabling semantic interoperability), i.e. very similar to REST. As such, it’s wrong to call IoT a technology.

So, if your nuclear power plant or a defense system is a closed network that operates in an isolated environment, it’s a distributed embedded system. If it’s open and scalable, homogeneous, reconfigurable, self-configurable and is taking advantage of machine learning, AI and data analytics (sensor based), it is IoT indeed!”

The examples of IoT

INDUSTRIAL INTERNET OF THINGS BASICS 2-3

Не повторяться с предыдущим параграфом

Выявить точное определение что такое интернет вещей в индустрии, перечислить в каких индустриях, какие обязательные составляющие идустриального интернета

First of all, as some important terms were already listed in this paper for the understanding of Industrial Internet of Things concept it s necessary to define the difference between IIoT, Industry 4.0, M2M and embedded systems. Some

– End-node / Device / Sensor / Thing etc.
– Gateway / Edge / Concentrator / Aggregator
– Cloud

yThe Internet of Things is simply “A network of Internet connected objects able to collect and exchange data.” It is commonly abbreviated as IoT.

The word “Internet of Things” has two main parts; Internet being the backbone of connectivity, and Things meaning objects / devices .

In a simple way to put it, You have “things” (things are nothing but your embedded system devices) that sense and collect data and send it to the internet. This data can be accessible by other “things” too.

Coming to your question, IoT = ET + NT + IT
confused?

At the simplest level, IoT is a combination of Embedded Technology (ET), Network Technology (NT) and Information Technology (IT)..

Let me give you a practical example. Lets imagine you have a “Smart air conditioning unit” in your home that is connected to the internet. (This is a “thing” connected to the internet) Now, imagine it’s a hot summer day and you have left for home from your work. You would like your home to be cool enough by the time you enter it. So, When you leave from your office, you can remotely switch ON the air conditioning unit of your home using your mobile (another “Thing” connected to the internet). Technically, with internet, you can control your AC system from any part of the world as long as both the AC and your mobile are connected using the “Internet”.

A further extension to this concept is: your mobile will command your home A/C that you are leaving the office (it can detect your GPS co-ordinates and decide your are on the move) and depending on the temperature, the A/C will be switched ON by your mobile itself, and the mobile will simply notify you that the A/C is ON.

your Smart A/C will have an embedded system that collects temperature data from a sensor and send it to the cloud (internet) using a wifi module. This is your Embedded system.

The Wifi network and Cloud constitute your Network Technology.

Your mobile will have an APP running in it that will receive the data. Depending on the received data, the app (in turn the mobile) will switch ON the A/C depending on your GPS co-ordinated. The mobile app infrastructure is a simple Information Technology.

So, all you need to do is, add NT & IT infrastructure to your Embedded system to convert it into an IoT system.

1.4k Views · View Upvotes

With the rise of Industrial Internet, businesses are focusing more and more on selling operational reliability instead of equipment to their customers. Machine learning is a branch of artificial intelligence, and a key driver in this development. Simon Jagers, the CEO of Semiotic Labs believes that ultimately artificial intelligence will improve the quality of life of all humans.

Machine learning means providing computers with the ability to learn from the inputs they have received. It could be described as enabling computers to predict patterns based on previous patterns, without having been explicitly programmed to do so. A computer receives inputs, learns which factors in the data are relevant when it comes to making predictions, and uses those learnings to improve its predictions about the influence of certain aspects.

“For example, a car battery running hot when driving at high speed is expected, but a hot battery hours after operating implies a short circuit. Machine learning enables the machine to determine which factors matter in different circumstances, and predict the outcome,” Jagers explains.

Fundamentally, machine learning helps detect anomalies. Another good example is the vibration analysis for wind turbines: Heavy vibration in a windy weather is a precondition, but when there is no wind, detecting heavy vibration is most likely a sign of an issue that needs to be addressed. In this case, there are only two factors that need to be considered. Nevertheless, with machine learning, up to thousands of factors and dimensions of data can be used to build predictions.

Ensuring operational reliability

Having the ability to spot anomalies at an early stage of a given process helps companies to focus on delivering operational reliability to their customers and users.

“Think of a modern rail transport system. If one railroad switch fails, it is not just one train that will not be able to pass it. A seemingly small error can cause severe problems and major delays in the entire network, which in addition can result in a lot of negative publicity and such”, Jagers says.

“Having the ability to spot anomalies at an early stage of a given process helps companies to focus on delivering operational reliability to their customers and users.”

In complex process industries, such as food production, impacts of a single failure in the system can reach both ends of the value chain. He continues: “A delay in the product distribution can lead to a lot of the so-called raw material, such as meat, going to waste, which of course is not only financially but also ethically and morally unacceptable. At the same time, failing to deliver on time has consequences on the retailers and their customers too. It is the reliability that matters – if one thing goes wrong, there are tons of other things that go wrong too.”

From time-based to condition-based maintenance

Machine learning does not only help prevent possible problems from occurring. It can also help optimize and streamline processes in the sense that it saves manufacturers and service providers from having to carry out certain procedures just in case. Equipment downtime is costly, so when it comes to maintenance, many want to be on the safe side.

“It is very hard to predict the exact time and reason for an equipment failure. On top of that, there is often a decreasing number of qualified maintenance technicians available. This often means having to increase efficiency when it comes to determining equipment’s need for repair and maintenance. In addition to helping detect anomalous functions in a particular equipment, machine learning can be utilized to create a list of all the pieces of equipment that need maintenance in a given facility, why they need it and when will they need it,” Jagers describes.

Having this kind of detailed information enables transferring from time-based to condition-based maintenance. “A lot of repair procedures are carried out before they are actually required. With condition-based maintenance, equipment is repaired exactly when needed, which brings more reliability, efficiency and cost-savings,” Jagers concludes.

Artificial intelligence as means of empowerment

Machine learning is an important driver in shifting towards pay-per-use models, where manufacturers focus on delivering operational reliability instead of pieces of equipment. Jagers, however, brings up a broader perspective on artificial intelligence and its impact on our future.

He sees computers as tools that can empower people to make better decisions and therefore become more productive: “I firmly believe that no matter how much artificial intelligence develops, computers can never fully replace human workers,” he ponders.

“Nevertheless, what they can do, is increase the overall productivity of the workforce by being capable of processing vast amounts of data and drawing conclusions from it. And that is really the most valuable advantage, because learning to do more with less is actually the only way of increasing prosperity in the world,” he summarizes.

In an IIoT system, many different new Information and Communication Technologies (ICTs), such as Industrial Wireless Networks (IWN) and Internet of Things (IoT) [16], [17] are incorporated into a single system. Similarly, in this prototype platform, some new ICTs are introduced. The architecture consists of four components: machines and equipment, networks, the cloud and terminals. Obviously, as shown in Figure 1, the prototype system is a closed loop for producing specific and personalized products to meet the users’ needs and desires. First of all, the users design the products according to their preference or provide the key parameters for personalized products through web pages. Then, the web server submits the user information to the industrial cloud, which parses the product data and the key parameters. Meanwhile, these optimization producing data are transmitted to industrial robots, workmen, and controllers of conveyor belts via wired or wireless networks. The production system begins to create the products depending on these data. During the manufacturing process of the product, all kinds of related data are transmitted to the cloud and neighboring nod

es for management and optimization.

On one hand, the plan not only provides the necessary information for managing and monitoring production, but also allows the optimization of processes and procedures for ensuring higher quality and increasing the production efficiency. On the other hand, the user can amend the design according to the manufacturing data. In the same way as before, these modifications and re-optimizations are delivered via all kinds of wired or wireless networks after being processed by the industrial cloud.

1) Physical Layer:

The physical layer is the basic component, and directly determines the specific type implementation and production. Meanwhile, from a functional perspective, it is responsible for the specific physical activities, such as manufacturing, transportation, mobility, logistics, and obtaining sensor or other data. In this platform, all kinds of devices, such as AGVs (Automated Guided Vehicles), manipulators, flexible conveyor systems, manufacturing equipment, warehouses, and sensors, may compose the physical layer. The following setup illustrates the working processes of the physical layer in this platform.

After obtaining the working instructions from the industrial cloud, the AGV first begins to carry the raw products with RFID tags from the warehouse to the entrance of a flexible conveyor system. The RFID tags include the key manufacturing information and the data of producing progress. Then, the raw products are transported to the corresponding manipulators, where machines and workmen prepare them for the ensuing processing. After that, the products are transported to the output exit by the conveyor belts, and the AGVs carry the finished products to the warehouse. During the loop, all the sensors have the ability to record the key parameters necessary for monitoring and alerting, and save the product’s information during the whole processing.

2) Networks:

Actually, for the smart factories of Industry 4.0, it is widely understood that wired or wireless networks must permeate the platform for transmitting data, commands and other information between the cloud and the equipment, including both machines and products. In all cases, networks play an important role in the implementation of Industry 4.0. In other words, networks play a role similar to the human body’s nervous system. In a similar manner, in the prototype framework of Industry 4.0, several network technologies are used to support the platform. The networks are formed by inter-factory networks, including IWNs, industrial Ethernet, NFC (Near Field Communication) utilizing RFIDs, MCNs (Mobile communication networks), civilian internet, etc.

From an intra-factory aspect, there are several types of data transmission taking place. Firstly, in order to determine and verify the location of products, RFID tags are mounted on the products. The RFID tags are used to read or write information about the products. Then, the RFID Gateway (e.g. a Raspberry Pi with Linux OS) transmits the product data using wireless protocols, in this case using a USB-WiFi module (IEEE 802.11), to send the related information to the equipment and to the cloud via access points (e.g., MOXA-AWK-3121). Secondly, in this system we provide wireless communications capabilities on equipment, such as the manufacturing machines and AGVs using technologies, such com-Wi-Fi (e.g., MOXA-Nport-W2150A) or wired-Wi-Fi (e.g., MOXA-AWK-1121). Then, the equipment can communicate with other equipment, the cloud, and the products via wireless networks because of the ease of access to civilian networks, such as the internet.

Outside the factory, there are two ways of dealing with the problem of communicating with users and management. A wired network is the best option for complicated technologies such as the internet. In the platform, we use Ethernet for connecting to users and inter-factory networks. With the development of mobile communications, especially 4G and 5G, mobile networks are becoming more widespread, so in order to provide easy access to this system for the users and the management, some mobile communication technologies can be introduced into the networks.

3) Industrial Cloud:

Based on the above discussion, it is evident that the cloud layer plays an important role in Industry 4.0, since the cloud not only performs computing for the optimization algorithms and decision making, but also stores massive data [18], [19]. Specifically, in this platform, the cloud is responsible for resolving the users’ needs for products, optimizing the flexible conveyor system, fusing and storing data, and even simple data mining operations.

The prototype platform consists of hardware and software. Five servers are used to construct the cloud. The machines were identical models NF8480M3 of Inspur Cor. with 8GB memory, a hard disk capacity of 500GB and 16 CPU cores each. The cloud setup used Citrix XenServer 6.5 and Apache Hadoop. The latter is a software library which acts as a framework that allows for the distributed processing of large data sets across clusters of computers using simple programming models. For storing data from the equipment, and the users, a MySQL database system was adopted in the platform.

4) Intelligent Terminals:

The intelligent terminals are directly used to display the related information and key data using web pages, messaging applications or emails. They can provide an effective, specialized, and visual way for users, workmen, and management to interact with the system. Large LCD screens, smart phones and PCs compose the terminal layers. Furthermore, the content’s structure can be adapted according to different aims, goals and classes.

Using user web page as an example, the terminals can be divided into two kinds of functions: reservation of resources and monitoring of the production process. In the reservation page, users can provide key parameters according to their needs by choosing some different options or submit designs. Meanwhile, taking the words too literally, reloading web page displays can provide updated information about the products, such as procuring progress, quality, and logistics.

B. Information Exchange

As mentioned above, all kinds of status data (e.g., equipment status data, product data, and measurement and control data) can be gathered by wired or wireless sensor nodes and forwarded to the industrial cloud platform. After analysis, macro-control of the devices may be realized in order to harmonize the different kinds of devices. In summation, the information exchange can be divided into three kinds: information exchange among physical layer devices, information exchange between physical layer devices and IWNs, and data processing in cloud and interaction through mobile terminals. Figure 2 shows the information exchange occurring in an industrial environment.

Information exchange among physical layer devices. The physical layer devices (e.g., AGVs, robotic arms, and conveyors) work together to process one or multiple products simultaneously. We need to develop a mechanism to support autonomous decision making and negotiations between smart entities. For example, when we simultaneously process multiple products, more machines may be immediately scheduled to support the completion of these tasks. If the number of assignments is gradually decreasing, the number of corresponding machines engaged can be automatically reduced by the coordination mechanism. As we can see, information exchange is ubiquitous among physical layer devices.
Information exchange between physical layer devices and the cloud. All the related data, such as equipment status data, are transferred to the cloud through IWNs. For the traditional methods, we should define the data format for the interaction protocol. For example, the RFID reader obtains the information of raw products, including product ID, product type, etc. This information must be packaged according to the protocol. The cloud platform receives these data and then unpacks and extracts the information for further analysis. From the perspective of implementation, we may adopt a web service (e.g., Tomcat+Java) to receive data. When the service receives the uploaded data, we may call the Flume interface to forward the data to distributed storage systems, such as HDFS, Hive, and HBase.
Data processing in the cloud and interaction with mobile terminals. We select HDFS and Hadoop as the uniform data storage and the foundation framework of the distributed system respectively. MapReduce is selected as the data processing method. The cluster resource management system, YARN, is used to coordinate the allocation of resources between Hive and Hbase. The processed data is exported to MySQL from Hive through the Sqoop interface. The interface server provides the information for external access through the communication protocol in the format of a HTTP interface. In this way, the customer can place an order, and the administrator has access to the real time information of the enterprise through mobile terminals.

PRECONDITIONS FOR IIOT IMPLEMENTATION 2-3

Перечисляю вновь индустрии, где можно применить

Далее по конкретным индустриям объясняю, как там можно использовать или уже используется ииот

Вывожу итог – диаграмму в чем плюсы и минусы от внедрения.

Industry 4.0 is part of the high-tech strategy of the German Federal Ministry for Education and Research [1], [2]. The term Industry 4.0 derived from the 4th industrial revolution and is a technology-oriented concept mainly for the manufacturing domain but can be interpreted more generally and applied to any value chain organization.

The 4^th industrial revolution means a paradigm shift in industry: The first industrial revolution brought the mechanization of production, the second industrial revolution was about mass production and the third industrial revolution means the digitization (electronic component, computer and IT. Industry 4.0 enables suppliers and manufacturers to leverage new technological concepts like CPS (Cyber-Physical Systems), Internet of Things and Cloud Computing (CC): New or enhanced products and services can be created, cost can be reduced and productivity can be increased [3].

In the discussions about Industry 4.0 the term Smart Factory [4] is mentioned quite often and even if the word smart might be open to debate, the main point is clear: The idea is to create a network with decentralized decisions and technological components with enhanced capabilities that can interact with each other and with human in real time. The components are (semi-) autonomous and can be equipped with advanced computing power or AI (Artificial Intelligence).

Industry 4.0 comprises the following terms or technological concepts:

Embedded Systems (ES) / CPS (Cyber-Physical Systems): Networks of IoT devices that interact physically with its environment, e.g. industrial robots with sensors and actors need physical input and provide physical output. ES and CPS are can also be equipped with digital interfaces. An ATM is an example for an embedded system.
Internet / Cloud of Things (IoT / CoT): Physical objects or components like ES or CPS that contain software and are connected to a network (Internet connectivity) and typically to a Cloud application creating opportunities for new services through integration of the physical and the digital world, e.g. automatic monitoring. The “things” can be any items, e.g. a conveyor belt in a factory. In the context of Industry 4.0 the term Industrial Internet of Things (IIoT) is used.
Service-Oriented Architecture (SOA) / Internet of Services (IoS) / Cloud Computing (CC): Service-oriented and cloud-based infrastructures and applications have advantages concerning scalability, elasticity, reliability, performance, device and location independence and more. Service models are for example Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS). Cloud applications can also be categorized in private, public or hybrid Cloud. Products for CC are available as open or closed source software systems [5], [6].

Important topics like Big Data and Business Intelligence are also part of Industry 4.0, but in order not to go beyond the scope of this short introduction, the concepts mentioned above are considered sufficient for the understanding of the idea of Industry 4.0.

The terms Internet of Things (IoT) and Cloud Computing (CC) imply a technology-oriented standpoint, e.g. devices and other electronic things play an active role in an environment and usage context that is connected (to the Internet). But when it comes to application scenarios, a sole technology-centric approach could be counterproductive: IoT devices are not used for their own sake; they must contribute to a more effective or a more efficient business process or a better user experience.

SECTION II.

 Business Process Modeling for Industry 4.0 Applications

Business process models (BPM) describe behavioral aspects of a system and are usually on the formal requirements or early design level. All major architectural and methodological frameworks include process modeling: TOGAF with the architecture domains (e.g. Business Architecture) and the ADM (Architecture Development Method) [7], RM-ODP (Reference Model of Open Distributed Processing) with its Enterprise and Computational Viewpoint [8], [9], and the FEAF (Federal Enterprise Architecture Framework) with its Business Reference Model [10]. In terms of OMG’s MDA (Model Driven Architecture) BPM is a PIM (Platform Independent Model) concerning the design and implementation platforms [11].

The notation language in this context is the OMG Business Process Model and Notation (BPMN 2.0) [12] with its modeling elements like tasks (service, send, receive etc.), events and data objects. Language extensions for aspects of Industry 4.0 application are needed, e.g. IoT-aware process models (IoT-A, Internet of Things Architecture [13], [14]). Existing extensions described in [13] and [15] are used and also a new extension (for Cloud App) are presented here. This BPMN-based language is called Industry 4.0 Process Modeling Language (I4PML): The following table shows the extensions as icons.

ROLE OF SENSORS IN IIOT 1-2

[15] [5] [16]

Сенсоры бываю такие и такие

To learn more about sensors and IIoT, see some of our favorite readings including:

How the Internet of Things Is Shaping the Sensor Market (Automation World)
Sensors are Fundamental to Industrial IoT (Automation.com)
Driving Unconventional Growth through the Industrial Internet of Things (Accenture)
3 Ways The Internet Of Things Will Change Every Business (Forbes)

SUMMARY

The Industrial Internet of Things being closely connected to the IoT is a

EVALUATION OF BIG DATA UTILIZATION IN IIOT

The section is devoted to answer a research question “How is Big Data linked to IIoT and what are its possible implications?” as well as issues of data collecting/storing/processing, the characteristics of big data and the current state of its appliance in terms of IIoT.

COEXISTENCE OF BIG DATA AND IIOT

Big data technologies create a lot of new opportunities and benefits for the industrial sector and at the same time they lead to great changes in the industry and cause the digital transformation.

Когда мы говорим о индустриальном интернете, то the data generated by industrial equipment “has a higher potential value than other types of big data” that we are used to for example different sources of consumer data.

Through predictive analytics, the ability to track the state of the infrastructure components IIOT significantly reduces costs on maintenance. According to the Accenture report [4] preventive maintenance reduces scheduled repairs by 12 percent, and thus costs on maintenance by 30 percent preventing 70 percent of shutdowns.

[4]

Sensorfusion,

DATA SOURCES FOR IIOT

Нужно выделить те области, где ииот

После этого здесь для разных сфер выписываем возможные ресурсы, можно нарисовать картинку, главное – учесть то, что не только сенсоры но и прочие каналы, написать про эмбедед системс и сайберсистемс

DATA GATHERING, STORING AND PROCESSING ISSUES

http://cdn.iotwf.com/resources/71/IoT_Reference_Model_White_Paper_June_4_2014.pdf

после того как есть ресурсы, пишем о том, как с них берутся данные, про сети передачи данных, про то о каких объемах речь и как они хранятся, особый уклон на неструктур в конце

IIOT ARCHITECTURE ASPECTS INFLUENCED BY BIG DATA

http://cdn.iotwf.com/resources/71/IoT_Reference_Model_White_Paper_June_4_2014.pdf

CLASSES OF IIOT SEVICE

пишем, что такое биг дата после чего примеры того, Как такие данве берутся из индестрии.

CASE STUDIES

EXISTING MARKET IIOT SOLUTIONS

WEB-SITES CONTENT ANALYSIS

This part will consider the existing IIoT solutions that can be divided into two cathegories. First one is listing the solutions of big well known companies without taking into account their reports where the detailed information can not be acquitted. The second set of companies includes startups that appeared for several past years and call themselves as IIoT providers.

The Table 1 shows the information about the companies that work with IIoT solutions describing the kind of service they provide. Considering the classes of IIoT services

Table 1

GE
AMAZON
ACCENTURE
SAP
IBM
MICROSOFT
ORACLE

EXPERT INTERVIEWS

CONCLUSION ON IIOT VISION ON THE MARKET

BIG DATA ENABLING IN IIOT

DEVELOPMENT OF THE THEORETICAL MODEL

https://iot.eclipse.org/resources/white-papers/Eclipse%20IoT%20White%20Paper%20-%20The%20Three%20Software%20Stacks%20Required%20for%20IoT%20Architectures.pdf

Классы сервисов иот по рребованиям [19]

http://guidingmetrics.com/content/cloud-services-industrys-10-most-critical-metrics/

Кждая компонента влияет на сервис

Сервис = ф(задержкаЮ частота обновления, аналитика и тд)

Response Time, Availability, Security, Analytics, Application Enablement, Connectivity, Edge Devices and Networks, Continuity

Continuity Continuity ensures that the service is available for certain amount of time without any interruptions. Furthermore if there is an incident, continuity allows to restart the service and regain access to data and functionality of the service within particular time frame. 2 Quality (of service) Quality describes attributes of service offering like access time, Availability Continuity Incident management Security Quality Monitoring Security Functionality Data access Security Service availability (in the clouds) Page 6 of 16 number of supported users, amount of data processed in a way that is convenient for the user for example without any lags or timeouts. 3 Functionality Functionality describes what end user can do and achieve in a particular time frame using selected services. 4 Incident management Incident management deals with incidents concerning services used by the end user. In case of any disruption of any above components the end user should be notified. 5 Monitoring In order to assure proper availability levels both the service owner and the service end user should be able to monitor it. It means that continuity, quality, functionality and security components should be measured and checked constantly. 6 Data access We can’t speak about availability if we forget the data either entered by the end user or data generated by the service based on end user data. Basically all the data end user enters and process including all results of this process should be available on-demand by the end user. 7 Security Obviously security is tightly connected with availability and its components like incident management, monitoring and data access. The end user expects that all those processes are in place and are working correctly. Furthermore as part of data access he expects his data to be secure i.e. no unauthorized access to his data is possible.

http://ec.europa.eu/justice/contract/files/expert_groups/discussion_paper_service_availability_en.pdf

https://secure.in.gov/iot/files/Service_Level_Agreement_-_FY15.pdf

http://www.cisco.com/c/dam/en/us/solutions/collateral/service-provider/mobile-internet/iotwf-whitepaper-transformation-next-generation-iot-sp.pdf

The Services Platform must surpass vertical solutions by integrating all essential technologies and required components into a common, open, and multi-application environment. The functions of the IoT Services Platform include the ability to deploy, configure, troubleshoot, secure, manage, and monitor IoT devices. They also include the ability to manage applications in terms of software/firmware installation, patching, starting/stopping, debugging, and monitoring. The Services Platform also provides capabilities that simplify application development through a core set of common application services that include data management, temporary caching, permanent storage, data normalization, policy-based access control, and exposure. In addition to these, the Services Platform is expected to offer some advanced application services, which include support for business rules, complex event processing, data analytics, and closed loop control. Figure 7.1 shows examples of key IoT Services Platform functions. A more detailed and structured list will be provided in Sects. 7.2–7.12. As can be seen from the list in Fig. 7.1, many of the capabilities of the IoT Services Platform represent what can be loosely categorized as “management functions.” These, however, are different from traditional network management. Traditional network-level management functions were originally defined, in the early 1980s, by the Open Systems Interconnection—Systems Management Overview (OSI-SMO) standard as FACPS: fault, configuration, accounting,

Traditional Management • Fault Management & Troubleshooting • Configuring & Deploying • Accounting & Billing • Performance Monitoring • Security Management Application Management • Software/firmware installation • Patching • Starting/stopping • Debugging • Monitoring Application Development • Data Management • Temporary Caching • Permanent Storage • Data Normalization, • Policy-based Access Control & Exposure Application Services • Business Rule Support • Complex Event Processing • Data Analytics • Closed Loop Control. • Subscriptions & Notifications • Service Discovery Fig. 7.1 Examples of key IoT Services Platform functions

performance, and security. A decade later, the Telecommunications Management Network (TMN) of ITU-T advanced the FCAPS functions as part of the TMN recommendation on management functions. The term FCAPS is often used in network management books as a useful way to break down the multipart network management functions. While FCAPS still apply, the overall management functions of IoT solutions are more multifaceted than traditional networks. This is due to the following factors: • IoT solutions include new devices (e.g., sensors, white-label gateways, and white-label switches). Some of these devices are inexpensive and generally lack the type or level of instrumentation required for traditional management functions. • IoT solutions utilize relatively recent technologies (e.g., tracking exact location of IoT device using GPS triangulation) that were not considered by traditional management solutions. • IoT solutions support more than two dozen access protocols (as was mentioned in Chaps. 4–5). The network management for each protocol may vary. • IoT solutions support multiple verticals, each of which has different sets of management, quality of service, and grade of service requirements. • IoT solutions utilize a new fog layer with new and challenging network, compute, and storage management requirements. • Finally, many enterprises and service providers are expected to outsource and, in many cases, multisource key parts of the network and/or management functions. This requires additional, mostly new, capabilities such as secure integration that spans connecting workflows between multiple service providers. This chapter describes the essential functions of the IoT Services Platform (Fig. 7.2). It focuses on identifying key capabilities with minimum emphasis on the relationship between the functions or their access protocol interfaces. Such relationship and protocols were addressed in the IoT Protocol Stack Chaps. 4 and 5.

OPPORTUNITIES FOR THE COMPANIES

POSSIBLE WAYS OF IMPLEMENTING IIOT IN COMBINATION WITH BIG DATA

PRECONDITIONS, REQUIREMENTS AND LIMITS

SUMMARY

DISCUSSION

RELATIONSHIP OF RESULTS TO THE RESEARCH QUESTION

GENERALIZATION OF RESULTS

HIGHLIGHT UNEXPECTED AND/OR INTERESTING FINDINGS

PUTTING THE RESULTS INTO THE CONTEXT OF THE RESEARCH AREA

DRAW CONCLUSIONS

CONCLUSION – SUMMARY – OUTLOOK

Expected results

Research Question 1 is devoted to defining terms IoT, IIoT, analyze the approaches of different companies to describing and implementing this idea and show the importance of IIoT in the modern world.

Research Question 2 will clarify the connection between Industrial Internet of Things and the term Big Data, including data gathering, storing, and processing techniques and describe the main components of the IIoT architecture when it works with Big Data.

Research Question 3, by mapping the definition of IIoT to cases of real applications in the industry, is expected to reveal that an amount of companies in the fields of industry abuse the term Big Data and IIoT by using them wrong. The research of the existing so-called IIoT solutions and expert interviews will help to conclude how companies understand the features of this concept and if they provide products that have necessary characteristics to be called IIoT and Big Data at the same time. In addition, the theoretical model will be created to show possible ways of implementing IIoT with Big Data in the industry, including issues of devices usage, gathering, storage and transferring data and the approaches for conducting the analysis.

Overview Research questions – methods – expected results

Research questions/ hypotheses

Method(s) per research question

Expected result(s) per method

What is the concrete definition of Industrial Internet of Things?

Literature review

Content analysis

Based on the literature review, it is planned to create the detailed definition of IIoT and list the main characteristics of its application.

Content analysis will provide with the information on the current state of IIoT and supplement for its defining based on the real examples.

How is Big Data linked to IIoT and what are its possible implications?

Literature review

Content analysis

Literature review will help to cr clarify the connection between Industrial Internet of Things and the term Big Data

Content analysis will be focused on the assessment of Big Data role in existing IIoT solutions

Which misunderstandings occur in the industry while applying Big Data IIoT solutions?

Expert interview

Case study

Modeling

Case studies and expert interviews will define and prove the real state of the Big Data implication in the existing IIoT solutions, how companies understand IIoT and if they provide products that have necessary characteristics to be called IIoT and Big Data at the same time

Modeling will help to create a theoretical model that shows possible Big Data implication in IIoT

Table of contents of the planned work

Technikum Wien Title Page

Declaration of authenticity

Abstract

Actuality

Problem Description

Expected Results

Table of contents

Introduction (10 %) 4

Formulation of the problem area, theoretical and practical purpose

Objectives, research questions

Announcement of the structure of the work;

Characteristics of the main information sources and methods used

Structure of the future work

Main body-parts:

Fundamentals (20 %) 8

Internet of Things

Industrial Internet of Things (IIoT)

Preconditions for IIoT implementation

Role of Sensors in IIoT

Big Data and IIoT

Evaluation of Big Data utilization in IIoT (40 %)16

Big Data in IIoT

Sources

Data gathering, storing and processing issues

IIoT architecture aspects influenced by Big Data

Case Studies

Existing market IIoT solutions

Web-sites content analysis

Expert interviews

Conclusion on IIoT vision on the market

Big Data enabling in IIoT (20 %) 8

Development of the Theoretical Model

Opportunities for the companies

Possible ways of implementing IIoT in combination with Big Data

Preconditions, requirements and limits

Discussion (5 %) 2

Relationship of results to the research question

Generalization of results

Highlight unexpected and/or interesting findings

Putting the results into the context of the research area

Draw conclusions

Conclusion – Summary – Outlook (5 %) 2

Research question, approach, theory, research area and methods with focus on the results

Outlook – What is possible further research?

Appendix and indexes

LITERATURE

[1]	D. O’Halloran and E. Kvochko, “Industrial Internet of Things: Unleashing the Potential of Connected Products and Services,” World Economic Forum, 2015. [Online]. Available: http://www3.weforum.org/docs/WEFUSA_IndustrialInternet_Report2015.pdf. [Accessed 24 Oct 2016].
[2]	P. C. Evans and M. Annunziata, “Industrial Internet: Pushing the Boundaries of Minds and Machines,” 26 Nov 2012. [Online]. Available: https://www.ge.com/docs/chapters/Industrial_Internet.pdf. [Accessed 27 Oct 2016].
[3]	K. Schwab, The Fourth Industrial Revolution, Davos-Klosters, Switzerland: World Economic Forum, 2016.
[4]	GeneralElectric; Accenture, “Industrial Internet Insights Report for 2015,” 2015. [Online]. Available: http://www.ge.com/digital/sites/default/files/industrial-internet-insights-report.pdf. [Accessed 27 Oct 2016].
[5]	P. Daugherty, P. Banerjee and W. Negm, “Driving Unconventional Growth through the Industrial Internet of Things,” 2015. [Online]. Available: https://www.accenture.com/us-en/_acnmedia/Accenture/next-gen/reassembling-industry/pdf/Accenture-Driving-Unconventional-Growth-through-IIoT.pdf. [Accessed 26 Oct 2016].
[6]	CBInsights, “The Industrial IoT: 56 Startups Transforming Factory Floors, Oil Fields, And Supply Chains,” CB Insights, 11 Jan 2016. [Online]. Available: https://www.cbinsights.com/blog/top-startups-iiot/. [Accessed 25 Oct 2016].
[7]	P. Teich, “Research Paper: Connecting with the Industrial Internet of Things (IIoT),” 29 Oct 2013. [Online]. Available: http://www.moorinsightsstrategy.com/wp-content/uploads/2013/10/Connecting-with-the-Industrial-Internet-of-Things-IIoT-by-Moor-Insights-Strategy.pdf. [Accessed 31 Oct 2016].
[8]	S. Dhanani, “Industrial Internet of Things (IIOT) and its Impact on the Design of Automation Systems,” 2014. [Online]. Available: http://pdfserv.maximintegrated.com/en/an/AN6142.pdf. [Accessed 30 Oct 2016].
[9]	“ITU-T. Y.2060 Next Generation Networks – Frameworks and functional architecture models,” International Telecommunication Union. Telecommunication Standardization Sector of ITU, 2012.
[10]	M. Ruggieri and H. Nikookar, “Internet of Things: From Research and Innovation to Market Deployment,” River Publishers Series in Communications, 2014.
[11]	O. Vermesan and P. Friess, “Internet of Things – Covering Technologies for Smart Environments and Integral Ecosystems,” River Publishers, River Publishers Series in Communications, 2013.
[12]	“NEAR FIELD COMMUNICATION,” [Online]. Available: http://nearfieldcommunication.org/. [Accessed 29 11 2016].
[13]	“MQTT.ORG,” [Online]. Available: http://mqtt.org/. [Accessed 29 11 2016].
[14]	T. Hänisch and V. P. Andelfinger, “Grundlagen: Das Internet der Dinge. Technik, Trends and Geschäftsmodelle,” Springer, 2015.
[15]	F. Mattern, “Die technische Basis für das Internet der Dinge,” Institut für Pervasive Computing, ETH Zürich, Zürich, 2006.
[16]	B. Marr, “3 Ways The Internet Of Things Will Change Every Business,” Forbes, 17 08 2015. [Online]. Available: http://www.forbes.com/sites/bernardmarr/2015/08/17/3-ways-the-internet-of-things-will-change-every-business/#292174ccd152. [Accessed 05 12 2016].
[17]	in Industrial IoT Technologies and Applications, International Conference, Industrial IoT 2016, 2016.
[18]	I. Ungurean and N.-C. Gaitan, “An IoT architecture for things from industrial environment,” Communications (COMM), 2014 10th International Conference on, 2014.
[19]	J. Soldatos, S. Gusmeroli, P. Malo and O. D. Giovanni, “Internet of Things Applications in Future Manufacturing,” Digitising the Industry – Internet of Things Connecting the Physical, Digital and Virtual Worlds. River Publishers Series in Communications, pp. 153-183, 2016.
[20]	Z. Sheng and C. Mahapatra, “Recent Advances in Industrial Wireless Sensor Networks Toward Efficient Management in IoT,” IEEE Access, 2015.
[21]	D. Evans, “The Internet of Things. How the Next Evolution of the Internet Is Changing Everything,” Cisco Internet Business Solutions Group (IBSG), 2011.
[22]	M. Diaz-Cacho and E. Delgado, “IoT integration on industrial environments,” Factory Communication Systems (WFCS), 2015 IEEE World Conference, Palma de Mallorca, 2015.
[23]	BuddeComm, “BuddeComm Intelligence Report – M2M, IoT and Big Data – Key Global Trends,” ThomsonOne © Copyright Paul Budde, 2015.
[24]	G. Brulte, “Grayson Brulte: The Industrial Internet Is Always Learning,” GE Reports, 8 Jul 2015. [Online]. Available: http://www.gereports.com/post/123475234518/grayson-brulte-the-industrial-internet-is-always-learnin/. [Accessed 28 Oct 2016].
[25]	S. Bangalore and S. Khandelwal, “4 Steps Towards Faster, Smarter Factories,” 28 Oct 2016. [Online]. Available: http://www.gereports.com/4-steps-towards-faster-smarter-factories/. [Accessed 30 Oct 2016].
[26]	“Industrial Internet of Things: Unleashing the Potential of Connected Products and Services,” World Economic Forum, Switzerland, 2015.

Explanation in which order the author is planning to solve the problems at hand;
Announcement of the structure of the work;

The main part consists out of chapters, which can be devided in to paragraphs, which in term can be devided in to items.

The first section, usually has a theorectical purpose, in which the author makes a critical analysis of the given material and draws their conclusions that would later be used in resolution of the given problems.
Material from monographs and journal entries and other sources is recomended to be be parapharsed and to be expressed in statistics and graphs.
In later sections the author should include the methods of solving the problems presented in the the text.
The contents of the practical part of the CPW must be concrete and objective, which means that they should be backed up by research and links to sources.
Each chapter of the work must be finished with a brief summary.
In the conclusion the author has to concetely formulate the results gathered during the development of the course project work and main conclusions that may be gathered from it.
The list of used liturature must be arranged in a certain order. It must include all the liturate used during the CPW, as well as the links, authors and their publication houses.
The appendix is not a nessesary element of the CPW. It is recommended when the author uses large statistics, graphs and charts.
In the CPW the author must maintain the scientific style of writing and terminology that is used in the jargon of the field. The author should avoid using sland and abbriviations. The author should avoid logical mistaxes, syntax and grammas mistakes.

A. IoT and Its Characteristics Research into the IoT is still in its early stage, and a standard definition of the IoT is not yet available. IoT can be viewed from three perspectives: 1) Internet-oriented; 2) things-oriented (sensors or smart things); and 3) semantic-oriented (knowledge) [6]. Also, the IoT can be viewed as either supporting consumers (human) or industrial applications and indeed could be named as the human Internet of Things (HIoT) or the industrial Internet of Things (IIoT) [19], [48]–[50]. Even though these different views have evolved because of the interdisciplinary nature of the subject, they are likely to intersect in an application domain to achieve the IoT’s goals. The first definition of the IoT was from a “things-oriented” perspective, where RFID tags were considered as things [6]. According to the RFID community, IoT can be defined as, “The worldwide network of interconnected objects uniquely addressable based on standard communication protocols” [51]. Fig. 1 illustrates the European research cluster of IoT (IERC) definition, where “The Internet of Things allows people and things to be connected anytime, anyplace, with anything and anyone, ideally using any path/network, and any service” [52], [53]. The International Telecommunication Union (ITU) views IoT very similarly: “From anytime, anyplace connectivity for anyone, we will now have connectivity for anything” [54]. Semantically, IoT means “A world-wide network of interconnected objects uniquely addressable, based on standard communication protocols” [51]. Most definitions of IoT do not explicitly highlight the industrial view of IoT (IIoT). World leading companies are giving special attention and making significant investments in the IoT for their industrial solutions (IIoT). Even though they use different terms such as “Smarter Planet” by IBM, “Internet of Everything” by Cisco and “Industrial Internet” by GE, their main objective is to use IoT to improve industrial production by reducing unplanned machine downtime and significantly reducing energy costs along with number of other potential benefits [19], [48]–[50], [55]. The IIoT refers to industrial objects, or “things,” instrumented with sensors, automatically communicating over a network, without human-to-human or human-tocomputer interaction, to exchange information and take intelligent decisions with the support of advanced analytics [50]. The definition of “things” in the IoT vision is very wide and includes a variety of physical elements. These include personal objects we carry around such as smart phones, tablets, and digital cameras. It also includes elements in our environments (e.g. home, vehicle, or work), industries (e.g., machines, motor, robot) as well as things fitted with tags (e.g., RFID), which become connected via a gateway device (e.g., a smart phone). Based on this view of “things,” an enormous number of devices will be connected to the Internet, each providing data and information, and some, even services. Sensor networks (SNs) including WSNs and wireless sensor and actuator networks (WSANs), RFID, M2M communications, and SCADA are the essential components of IoT. As described in more detail in this section, a number of the IoT’s characteristics are inherited from one or more of these components. For instance, “resource-constrained” is inherited from RFID and SNs, and “intelligence” is inherited from WSNs and M2M. Other characteristics (e.g., ultra-large-scale network, spontaneous interactions) are specific to the IoT. The main characteristics of the IoT are presented from infrastructure and application perspectives. 1) Characteristics of IoT Infrastructure: • Heterogeneous devices: The embedded and sensor computing nature of many IoT devices means that low-cost computing platforms are likely to be used. In fact, to minimize the impact of such devices on the environment and energy consumption, low-power radios are likely to be used for connection to the Internet. Such lowpower radios do not use WiFi, or well-established cellular network technologies. However, the IoT will not be composed only of embedded devices and sensors, it will also need higher-order computing devices to perform heavier duty tasks (routing, switching, data processing, etc.). Device heterogeneity emerges not only from differences in capacity and features, but also for other reasons including multivendor products and application requirements. [4], [54]. Fig. 2 illustrates six different types of IoT devices. • Resource-constrained: Embedded computing and sensors need a small device form factor, which limits their processing, memory, and communication capacity. As shown in Fig. 2, resource capacity (e.g., computational, connectivity capabilities, and memory requirements) decreases moving from left to right. For example, RFID devices or 72 IEEE INTERNET OF THINGS JOURNAL, VOL. 3, NO. 1, FEBRUARY 2016 Fig. 2. Examples of device heterogeneity in IoT. tags (in the right-most side of this figure) may not have any processing capacity or even battery to power them. On the other hand, in Fig. 2, devices become expensive and larger in form-factor when moving to the left. • Spontaneous interaction: In IoT applications, sudden interactions can take place as objects move around, and come into other objects’ communication range, leading to the spontaneous generation of events. For instance, a smartphone user can come in close contact with a TV/fridge/washing machine at home and that can generate events without the user’s involvement. Typically, in IoT, an interaction with an object means that an event is generated and is pushed to the system without much human attention. • Ultra-large-scale network and large number of events: In an IoT environment, thousands of devices or things may interact with each other even in one local place (e.g., in a building, supermarket, and university), which is much larger scale than most conventional networking systems. Globally, the IoT will be an ultra-large-scale network containing nodes in the scale of billions and even in trillions. Gartner has predicted [56] that there will be nearly 26 billion devices on the IoT by 2020. Similarly, ABI research [57] estimated that more than 30 billion devices will be wirelessly connected by 2020. In the IoT, spontaneous interactions among an ultra large number of things or devices will produce an enormous number of events as normal behavior. This uncontrolled number of events may cause problems such as event congestion and reduced event processing capability. • Dynamic network and no infrastructure: As shown in Fig. 2, IoT will integrate devices, many of which will be mobile, wirelessly connected, and resource constrained. Mobile nodes within the network leave or join anytime they want. Also, nodes can be disconnected due to poor wireless links or battery shortage. These factors will make the network in IoT highly dynamic. Within such an ad hoc environment, where there is limited or no connection to a fixed infrastructure, it will be difficult to maintain a stable network to support many application scenarios that depend on the IoT. Nodes will need to cooperate to keep the network connected and active. • Context-aware: Context is key in the IoT and its applications. A large number of sensors will generate large amounts of data, which will not have any value unless it is analyzed, interpreted, and understood. Context-aware computing stores context information related to sensor data, easing its interpretation. Context-awareness (especially in temporal and spatial context) plays a vital role in the adaptive and autonomous behavior of the things in the IoT [20], [58]. Such behavior will help to eliminate human-centric mediation in the IoT, which ultimately makes it easier to perform M2M communication, a core element of the IoT’s vision. • Intelligence: According to Intel’s IoT vision, intelligent devices or things and intelligent systems of systems are the two key elements of IoT [59]. In IoT’s dynamic and open network, these intelligent entities along with other entities such as Web services (WSs), SOA components, and virtual objects will be interoperable and able to act independently based on the context, circumstances, or environments [60], [61]. • Location-aware: Location or spatial information about things (objects) or sensors in IoT is critical, as location plays a vital role in context-aware computing. In a largescale network of things, interactions are highly dependent on their locations, their surroundings, and presence of other entities (e.g., things and people). • Distributed: The traditional Internet itself is a globally distributed network, and so also is the IoT. The strong spatial dimension within the IoT makes the network IoT distributed at different scales (i.e., both globally like the Internet, and also locally within an application area). 2) Characteristics of IoT Applications: • Diverse applications: The IoT can offer its services to a large number of applications in numerous domains and environments. These domains and environments can be grouped into (nonexhaustive) domain categories such as: 1) transportation and logistics; 2) healthcare; 3) smart environment (home, office, and plant); 4) industrial; and 5) personal and social domain. Fig. 3 highlights some key application domains for the IoT. Different applications are likely to need different deployment architectures (e.g., event-driven and time-driven) and have different requirements. However, since the IoT is connected to the Internet, most of the devices comprising IoT services will need to operate within an environment that supports their mutual understanding. • Real time: Applications using the IoT can be broadly classified as real time and non-real time. For instance, IoT for healthcare, transportation will need on-time delivery of their data or service. Delayed delivery of data can make the application or service useless and even dangerous in mission critical applications. RAZZAQUE et al.: MIDDLEWARE FOR IoT 73 Fig. 3. Potential applications of IoT [66]. • Everything-as-a-service (XaaS): An everything-as-aservice model is very efficient, scalable, and easy to use [62]. The XaaS model has inspired the sensing as a service approach in WSNs [63], [64], and this may inevitably lead IoT toward an XaaS model. As more things get connected, the collection of services is also likely to grow, and as they become accessible online, they will be available for use and reuse. • Increased security attack-surface: While there is huge potential for the IoT in different domains, there are also concerns for the security of applications and networks. The IoT needs global connectivity and accessibility, which means that anyone can access it anytime and anyway. This tremendously increases the attack surfaces for the IoT’s applications and networks. The inherent complexity of the IoT further complicates the design and deployment of efficient, interoperable, and scalable security mechanisms. • Privacy leakage: Using the IoT, applications may collect information about people’s daily activities. As information reflecting such activities (e.g., travel routes, buying habits, and daily energy usage) is considered by many individuals as private, exposure of this information could impact the privacy of those individuals. The use of cloud computing makes the problem of privacy leakage even worse. Any IoT application not compliant with privacy requirements could be prohibited by law (e.g., in the EU [65]) because they violate citizens’ privacy. B. Middleware in IoT and Its Requirements Generally, a middleware abstracts the complexities of the system or hardware, allowing the application developer to focus all his effort on the task to be solved, without the distraction of orthogonal concerns at the system or hardware level [67]. Such complexities may be related to communication concerns or to more general computation. A middleware provides a software layer between applications, the operating system and the network communications layers, which facilitates and coordinates some aspect of cooperative processing. From the computing perspective, a middleware provides a layer between application software and system software. In the IoT, there is likely to be considerable heterogeneity in both the communication technologies in use, and also the system level technologies, and a middleware should support both perspectives as necessary. Based on previously described characteristics of the IoT’s infrastructure and the applications that depend on it, a set of requirements for a middleware to support the IoT is outlined. As follows, these requirements are grouped into two sets: 1) the services such a middleware should provide and 2) the system architecture should support. 1) Middleware Service Requirements: Middleware service requirements for the IoT can be categorized as both functional and nonfunctional. Functional requirements capture the services or functions (e.g., abstractions, resource management) a middleware provides and nonfunctional requirements (e.g., reliability, security, and availability) capture QoS support or performance issues. The view of a middleware in this paper is one which provides common or generic services to multiple different application domains. In this section, no attempt is made to capture domain or application-specific requirements, as the focus is on generic or common functional ones, as follows. • Resource discovery: IoT resources include heterogeneous hardware devices (e.g., RFID tags, sensors, sensor mote, and smartphones), devices’ power and memory, analogue to digital converter devices (A/D), the communications module available on those devices, and infrastructural or network level information (e.g., network topology and protocols), and the services provided by these devices. Assumptions related to global and deterministic knowledge of these resources’ availability are invalid, as the IoT’s infrastructure and environment is dynamic. By necessity, human intervention for resource discovery is infeasible, and therefore, an important requirement for resource discovery is that it be automated. Importantly, when there is no infrastructure network, every device must announce its presence and the resources it offers. This is a different model to centralized distributed systems, where resource publication, discovery, and communication are generally managed by a dedicated server. Discovery mechanisms also need to scale well, and there should be efficient distribution of discovery load, given the IoT’s composition of resource-constrained devices. • Resource management: An acceptable QoS is expected for all applications, and in an environment where resources that impact on QoS are constrained, such as the IoT, it is important that applications are provided with a service that manages those resources. This means that resource usage should be monitored, resources allocated or provisioned in a fair manner, and resource con- flicts resolved. In IoT architectures, especially in serviceoriented or virtual machine (VM)-based architectures, middleware needs to facilitate potentially spontaneous resource (service) (re)composition, to satisfy application needs. • Data management: Data are key in IoT applications. In the IoT, data refer mainly to sensed data or any network infrastructure information of interest to applications. An IoT middleware needs to provide data management 74 IEEE INTERNET OF THINGS JOURNAL, VOL. 3, NO. 1, FEBRUARY 2016 services to applications, including data acquisition, data processing (including preprocessing), and data storage. Preprocessing may include data filtering, data compression, and data aggregation. • Event management: There are potentially a massive number of events generated in IoT applications, which should be managed as an integral part of an IoT middleware. Event management transforms simple observed events into meaningful events. It should provide real-time analysis of high-velocity data so that downstream applications are driven by accurate, real-time information, and intelligence. • Code management: Deploying code in an IoT environment is challenging, and should be directly supported by the middleware. In particular, code allocation and code migration services are required. Code allocation selects the set of devices or sensor nodes to be used to accomplish a user or application level task. Code migration transfers one node/device’s code to another one, potentially reprogramming nodes in the network. Using code migration services, code is portable, which enables data computation to be relocated. Key nonfunctional requirements of IoT middleware are as follows. • Scalability: An IoT middleware needs to be scalable to accommodate growth in the IoT’s network and applications/services. Considering the size of the IoT’s network, IPv6 is a very scalable solution for addressability, as it can deal with a huge number of things that need to be included in the IoT [68]. Loose coupling and/or virtualization in middleware is useful in improving scalability, especially application and service level scalability, by hiding the complexity of the underlying hardware or service logic and implementation. • Real time or timeliness: A middleware must provide realtime services when the correctness of an operation that supports depends not only on its logical correctness but also on the time in which it is performed. As the IoT will deal with many real-time applications (e.g., transportation, healthcare), on-time delivery of information or services in those applications is critical. Delayed information or services in such applications can make the system useless and even dangerous. • Reliability: A middleware should remain operational for the duration of a mission, even in the presence of failures. The middleware’s reliability ultimately helps in achieving system level reliability. Every component or service in a middleware needs to be reliable to achieve overall reliability, which includes communication, data, technologies, and devices from all layers. • Availability: A middleware supporting an IoT’s applications, especially mission critical ones, must be available, or appear available, at all times. Even if there is a failure somewhere in the system, its recovery time and failure frequency must be small enough to achieve the desired availability. The reliability and availability requirements should work together to ensure the highest fault tolerance required from an application. • Security and privacy: Security is critical to the operation of IoT. In IoT middleware, security needs to be considered in all the functional and nonfunctional blocks including the user level application. Context-awareness in middleware may disclose personal information (e.g., the location of an object or a person). Like security, every block of middleware, which uses personal information, needs to preserve the owner’s privacy. • Ease-of-deployment: Since an IoT middleware (or more likely, updates to the middleware) is typically deployed by the user (or owner of the device), deployment should not require expert knowledge or support. Complicated installation and setup procedures must be avoided. • Popularity: An IoT middleware (like any other software solution) should be continuously supported and extended. Usually, this facility is provided within a community of developers and researchers. While this is not necessarily a requirement, a large number of users who adopt a particular technology motivates future testing and development. 2) Architectural Requirements: The architectural requirements included in this section are designed to support application developers. They include requirements for programming abstractions, and other implementation-level concerns. • Programming abstraction: Providing an API for application developers is an important functional requirement for any middleware. For the application or service developer, high-level programming interfaces need to isolate the development of the applications or services from the operations provided by the underlying, heterogeneous IoT infrastructures. The level of abstraction, the programming paradigm, and the interface type all need to be considered when defining an API. The level of abstraction refers to how the application developer views the system (e.g., individual node/device level, system level). The programming paradigm (e.g., publish/subscribe) deals with the model for developing or programming the applications or services. The interface type defines the style of the programming interface. For instance, descriptive interfaces offer SQL-like languages for data query [69], XML-based specification files for context configuration [70]. • Interoperable: A middleware should work with heterogeneous devices/technologies/applications, without additional effort from the application or service developer. Heterogeneous components must be able to exchange data and services. Interoperability in a middleware can be viewed from network, syntactic, and semantic perspectives, each of which must be catered for in an IoT. A network should exchange information across different networks, potentially using different communication technologies. Syntactic interoperation should allow for heterogeneous formatting and encoding structures of any exchanged information or service. Semantic interoperability refers to the meaning of information or a service, and should allow for interchange between the evergrowing and changing set of devices and services in IoT. Meaningful information about services will be useful for the users in composing multiple services as semantic RAZZAQUE et al.: MIDDLEWARE FOR IoT 75 Fig. 4. Relationships between the IoT applications and infrastructure and its middleware requirements. data can be better understood by “things” and humans compared to traditional protocol descriptions [71], [72]. • Service-based: A middleware architecture should be service-based to offer high flexibility when new and advanced functions need to be added to an IoT’s middleware. A service-based middleware provides abstractions for the complex underlying hardware through a set of services (e.g., data management, reliability, security) needed by applications. All these and other advanced services can be designed, implemented, and integrated in a service-based framework to deliver a flexible and easy environment for application development. • Adaptive: A middleware needs to be adaptive so that it can evolve to fit itself into changes in its environment or circumstances. In the IoT, the network and its environment are likely to change frequently. In addition, application-level demands or context are also likely to change frequently. To ensure user satisfaction and effectiveness of the IoT, a middleware needs to dynamically adapt or adjust itself to fit all such variations. • Context-aware: Context-awareness is a key requirement in building adaptive systems and also in establishing value from sensed data. The IoT’s middleware architecture needs to be aware of the context of users, devices, and the environment and use these for effective and essential services’ offerings to users. • Autonomous: It means self-governed. Devices/ technologies/applications are active participants in the IoT’s processes and they should be enabled to interact and communicate among themselves without direct human intervention [5], [73]. Use of intelligence including autonomous agents, embedded intelligence [74], predictive, and proactive approaches (e.g., a prediction engine) in middleware can fulfil this requirement [75]. • Distributed: A large-scale IoT system’s applications/devices/users (e.g., WSNs and vehicular ad hoc networks) exchange information and collaborate with each other. Such applications/devices/users are likely to be geographically distributed, and so a centralized view or middleware implementation will not be sufficient to support many distributed services or applications. A middleware implementation needs to support functions that are distributed across the physical infrastructure of the IoT. Fig. 4 presents the relationships between the IoT’s middleware requirements and its infrastructural and application characteristics. As shown in this figure, most of the requirements are directly related (red colour text) to one or more characteristics of the IoT. A few of them are also indirectly linked (black text) to one or more characteristics of the IoT. For instance, the realtime behavior requirement is directly related to the application’s real-time characteristics and indirectly to the large number of events. Also, a few of the middleware requirements (e.g., resource discovery and resource management) jointly capture the same set of IoT characteristics.

Research into the IoT is still in its early stage, and a standard definition of the IoT is not yet available. IoT can be viewed from three perspectives: 1) Internet-oriented; 2) things-oriented (sensors or smart things); and 3) semantic-oriented (knowledge) [6]. Also, the IoT can be viewed as either supporting consumers (human) or industrial applications and indeed could be named as the human Internet of Things (HIoT) or the industrial Internet of Things (IIoT) [H. Zhou, The Internet of Things in the Cloud: A Middleware Perspective, 1st ed. Boca Raton, FL, USA: CRC, 2012.], [Moor Insights and Strategy. (2013). “Behaviorally segmenting the Internet of Things (IoT),” [Online]. Available: http://www. moorinsightsstrategy.com/wp -content/ uploads/ 2013/ 10/BehaviorallySegmenting-the-IoT-by-Moor-Insights-Strategy.pdf [49] C. P. Greg Gorbach and A. Chatha. (2014). “Planning for the industrial Internet of Things,” [Online]. Available: http://www.arcweb.com/ brochures/planning-for-the-industrial-internet-of-things.pdf [50] M. Scott and R. Whitney. (2014). “The industrial Internet of Things,” [Online]. Available: http://www.mcrockcapital.com/uploads/1/0/9/6/ 10961847/mcrock_industrial_internet_of_things_report_2014.pdf]. Even though these different

views have evolved because of the interdisciplinary nature of the subject, they are likely to intersect in an application domain to achieve the IoT’s goals

Share this: Facebook Twitter Reddit LinkedIn WhatsApp

Cite This Work

To export a reference to this article please select a referencing stye below:

Related Services

View all

Dissertation Writing Service

From £136

Dissertation Proposal Writing Service

From £124

Female student reading and using laptop to study

Topics and Titles Writing Service

From £24

DMCA / Removal Request

If you are the original writer of this dissertation and no longer wish to have your work published on the UKDiss.com website then please:

The Industrial Internet of Things (Industrial IoT) is made up of a multitude of devices connected by communications software. The resulting systems, and even the individual devices that comprise it, can monitor, collect, exchange, analyze, and instantly act on information to intelligently change their behavior or their environment — all without human intervention.	https://www.rti.com/industries/iot-faq.html

Dissertation Services

PhD Services

Other Services

Contact

CONTENTS

INTRODUCTION

Research Questions

STRUCTURE OF THE WORK

FUNDAMENTALS 7-11

INTERNET OF THINGS AS ORIGIN 1-2

INDUSTRIAL INTERNET OF THINGS BASICS 2-3

Related QuestionsMore Answers Below

Ensuring operational reliability

“Having the ability to spot anomalies at an early stage of a given process helps companies to focus on delivering operational reliability to their customers and users.”

From time-based to condition-based maintenance

Artificial intelligence as means of empowerment

1) Physical Layer:

2) Networks:

3) Industrial Cloud:

4) Intelligent Terminals:

B. Information Exchange

PRECONDITIONS FOR IIOT IMPLEMENTATION 2-3

 Business Process Modeling for Industry 4.0 Applications

ROLE OF SENSORS IN IIOT 1-2

SUMMARY

EVALUATION OF BIG DATA UTILIZATION IN IIOT

COEXISTENCE OF BIG DATA AND IIOT

DATA SOURCES FOR IIOT

DATA GATHERING, STORING AND PROCESSING ISSUES

IIOT ARCHITECTURE ASPECTS INFLUENCED BY BIG DATA

CASE STUDIES

EXISTING MARKET IIOT SOLUTIONS

WEB-SITES CONTENT ANALYSIS

EXPERT INTERVIEWS

CONCLUSION ON IIOT VISION ON THE MARKET

BIG DATA ENABLING IN IIOT

DEVELOPMENT OF THE THEORETICAL MODEL

OPPORTUNITIES FOR THE COMPANIES

POSSIBLE WAYS OF IMPLEMENTING IIOT IN COMBINATION WITH BIG DATA

PRECONDITIONS, REQUIREMENTS AND LIMITS

SUMMARY

DISCUSSION

RELATIONSHIP OF RESULTS TO THE RESEARCH QUESTION

GENERALIZATION OF RESULTS

HIGHLIGHT UNEXPECTED AND/OR INTERESTING FINDINGS

PUTTING THE RESULTS INTO THE CONTEXT OF THE RESEARCH AREA

DRAW CONCLUSIONS

CONCLUSION – SUMMARY – OUTLOOK

LITERATURE

Cite This Work

Related Services

DMCA / Removal Request