Electronics in advanced research industries: industry 4.0 to industry 5.0 advances alessandro massar
Visit to download the full and correct content document: https://ebookmass.com/product/electronics-in-advanced-research-industries-industry4-0-to-industry-5-0-advances-alessandro-massaro/
More products digital (pdf, epub, mobi) instant download maybe you interests ...
Implementing Industry 4.0 in SMEs. Concepts, Examples and Applications Dominik T. Matt
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
The right of Alessandro Massaro to be identified as the author of this work has been asserted in accordance with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty
While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging‐in‐Publication Data
Names: Massaro, Alessandro, 1974– author.
Title: Electronics in advanced research industries : industry 4.0 to industry 5.0 advances / Alessandro Massaro, Dyrecta Lab, Research Institute, Conversano (Ba), Italy.
Description: Hoboken, NJ, USA : Wiley, 2022. | Includes bibliographical references and index.
Identifiers: LCCN 2021028944 (print) | LCCN 2021028945 (ebook) | ISBN 9781119716877 (hardback) | ISBN 9781119716884 (adobe pdf) | ISBN 9781119716891 (epub)
Set in 9.5/12.5pt STIXTwoText by Straive, Chennai, India
To my family: Magda, Andrea, Adriano, and Peggy
Contents
Preface xiii
About the Author xv
1 State of the Art and Technology Innovation 1
1.1 State of the Art of Flexible Technologies in Industry 2
1.1.1 Sensors and Actuators Layer: I/O Layer 3
1.1.2 Agent/Firmware Layer: User Interface Layer 9
1.1.3 Gateway and Enterprise Service Bus Layer 9
1.1.4 IoT Middleware 10
1.1.5 Processing Layer 11
1.1.6 Application Layer 11
1.1.7 File Transfer Protocols 11
1.2 State of the Art of Scientific Approaches Oriented on Process Control and Automatisms 14
1.2.1 Architectures Integrating AI 14
1.2.2 AI Supervised and Unsupersived Algorithms 15
1.2.3 AI Image Processing 18
1.2.4 Production Process Mapping 20
1.2.5 Technologies of Industry 4.0 and Industry 5.0: Interconnection and Main Limits 21
1.2.6 Infrared Thermography in Monitoring Process 26
1.2.7 Key Parameters in Supply Chain and AI Improving Manufacturing Processes 27
1.3 Intelligent Automatic Systems in Industries 30
1.4 Technological Approaches to Transform the Production in Auto-Adaptive Control and Actuation Systems 31
1.5 Basic Concepts of Artificial Intelligence 33
1.6 Knowledge Upgrading in Industries 41 References 45
2 Information Technology Infrastructures Suppor ting Industry 5.0 Facilities 51
2.1 Production Process Simulation and Object Design Approaches 52
2.1.1 Object Design of a Data Mining Algorithm: Block Functions and Parameter Setting 55
2.1.2 Example 1: BPM Modeling of Wheat Storage Process for Pasta Production 59
2.1.3 Example 2: Block Diagram Design of a Servo Valve Control and Actuation System 61
2.1.4 Example 3: Block Diagram of a Liquid Production System 61
2.1.5 Example 4: UML Design of a Programmable Logic Controller System 62
2.1.6 Example 5: Electronic Logic Timing Diagram 64
2.1.7 Example 6: AR System in Kitchen Production Process 64
2.1.8 Example 7: Intelligent Canned Food Production Line 70
2.2 Electronic Logic Design Oriented on Information Infrastructure of Industry 5.0 71
2.3 Predictive Maintenance: Artificial Intelligence Failure Predictions and Information Infrastructure Layout in the Temperature Monitoring Process 74
2.4 Defect Estimation and Prediction by Artificial Neural Network 77
2.4.1 Other Methodologies to Map and Read Production Failures and Defects 79
2.5 Defect Clustering and Classification: Combined Use of the K-Means Algorithm with Infrared Thermography for Predictive Maintenance 82
2.6 Facilities of a Prototype Network Implementing Advanced Technology: Example of an Advanced Platform Suitable for Industry 5.0 Integrating Predictive Maintenance 84
2.7 Predictive Maintenance Approaches 86
2.7.1 Preventive Maintenance and Predictive Maintenance Operations in the Railway Industry 90
2.8 Examples of Advanced Infrastructures Implementing AI 93
2.9 Examples of Telemedicine Platforms Integrating Advanced Facilities 94
9.1.1 Photonic Crystal Pillars for Filtering and Optical Resonance 382
9.1.2 Thin Film Microelectromechanical System Prototyping and Photolithography Approach 387
9.1.3 Thin Film GHz Microstructures by the Photolithography Approach 387
9.1.4 Gas Sensing Homemade Experimental Setup for Rapid Prototyping 390
9.2 Examples of Antenna and Detection System Rapid Prototyping 392
9.2.1 GPR Antenna Design for UAV Integration System 392
9.2.2 Example of an Underground Water Leakage Detection System Integrating GPR, UAV, and Infrared Thermal Imaging: System Prototyping 397
9.2.3 Integrated Diamond Patch-Type Antennas and Applications 400
9.3 Principles of Mechanical Piece Rapid Prototyping and Innovative Materials 411
9.3.1 Example of Diamond Material Implementations 413
9.4 Rapid Prototyping and Artificial Intelligence Upgrade 415
9.5 Rapid Prototyping Oriented Toward Patent Development 418
9.5.1 Prototyping of Devices Implementing Nanoparticles 418
9.5.2 Prototyping of an Optoelectronic Device Based on a Nanocomposite Tip 418
9.5.3 DNA Lab-on-Chip 418
9.6 Nanocomposite Artificial Skin Rapid Prototyping Process 437 References 439
10 Scientific Research in Industry 445
10.1 Guidelines to Construct an Advanced Research Unit in Industry in the Electronic and Mechatronic Field 446
10.2 Guidelines to Formulate a Patent 448
10.3 Guideline to Propose Technological Advances for Public Entities and in Industry 5.0 Research Project 449
10.3.1 Setting of a Research Project of Underground Water Leakage 449
10.3.2 Setting a Research Project Involving Technologies for Hydrogeological Risk Monitoring 456
10.3.3 Setting a Research Project in Mechatronics: Production of a Diagnostic Machine by Means of Industry 5.0 Facilities 468
10.4 Innovation Process Projects: Example of a Smart Wine Factory 483
10.5 Guideline for Project Management 485 References 506
Abbreviations and Acronyms 507 Index 515
Modern technologies in production systems open new approaches and concepts of industrial production. The digital Industry 4.0 upgrade provides new elements to control and manage production in all industry sectors. This upgrade allows to improve product quality, and in general the whole supply chain. The new digital technologies include hardware and software tools integrated in infrastructure oriented on the gain of digital knowledge. The fast dynamicity of the markets, the increase of the global competition between companies, and the unpredictable social and health events, imposes the need to think of a new concept of a production system based on full automatisms and self-adaptive processes, predicting production failures and product defects. In this context, the Industry 4.0 facilities can be furthermore upscaled to an intelligent control and actuation system of the production, characterizing the new Industry 5.0 scenario. The new facilities which contribute to Industry 5.0 passage are mainly based on artificial intelligence (AI) implementations in production and information systems, accomplishing predictive maintenance, failure prediction, defect classification, efficient robotic control and actuation, design optimization, testing improvements, and in general technological advances due to the possibility to quickly process data in each production stage. This book analyzes innovative production approaches, and the integration aspects of the AI in different industrial digital technologies, by enhancing specific functionalities. In innovative production systems, AI is fully integrated in information systems and covers cybersecurity, quality processes, business intelligence and intelligent production management. The innovative production is also related to new services associated with the introduction in the market of new technologies such as for the telemedicine sector, and in general for industrial diagnostics, where AI is also adopted for the improvement of inspection services. The main advantage of AI is the self-learning of the algorithms able to learn automatically from the same production data of companies. In an industrial upgrade, the implementation of sensor control and actuation based on intelligent feedback systems is especially important. In this scenario, AI algorithms can accomplish robotic movement, by automatically optimizing the machine parameter setting, by means of image and data processing. The correct use of AI is mainly based on the formulation of the algorithm, and on the dataset adopted to learn the related model. For each application there is an associated AI learning dataset which can be improved by big data systems. In particular, image processing and image segmentation approaches can be improved by AI, enhancing hidden information as defects, or adding new information about the performed production process. Another tool supporting the assembly in supply chains and the coordination of activities is augmented reality, which can be fully integrated in the information company infrastructure. An important step for a new concept of production is the upgrade of the information technology (IT) infrastructure automatically gaining the knowledge. Different IT architectures are proposed for different application Preface
fields to enhance technologies more suitable for a self-adaptive production providing decision support systems. A particular interesting topic for the innovative IT is the Internet of Things. The design and the development of an advanced IT infrastructure is the primary action to add for the upgrade in Industry 5.0. The AI concept is extended to the logic condition implementation, acting on signal processing, and on the use of simply electronic circuits representing these logics. The discussed methodologies allow to comprehend how it is possible to move on a competitive production based on the concept of “flexible” production and on new products based on advanced technologies on a micro- and nanoscale. In this scenario, companies working in manufacturing can switch dynamically the production on the new products, thus converting the production in innovative components, machines, materials, sensors, or devices. Following this orientation, this book proposes important approaches to automatize efficiently the new production, by analyzing highly advanced production tools based on nanotechnology. In this direction useful methodologies are analyzed to implement the production of high technology devices, such as reverse engineering and rapid prototyping by showing different examples useful to comprehend the methodologies to apply for an innovative production based on scientific and industrial research development. Particular attention is paid to the procedures to follow to produce a new device, to increase the company capacity to accelerate the industrialization process starting a new innovative prototype fabrication, and to basic approaches for the design modeling and testing. The optimization of the preindustrialization process to perform is accompanied by a quick check of the basic properties of the new product to fabricate, and by the simultaneous support of the AI application improving analysis. In order to start and to develop a new research activity, also concerning advanced technologies, precise schemes must be followed. The discussed topics facilitate the understanding of the directions of the research for the production upscaling, just to apply the research activity. The last part of the book provides different elements useful for writing an industrial research project, and for the project management. This book deals with multidisciplinary topics including electronics, mechatronics, mechanics, and informatics. All the analyzed topics are useful to know; the key elements are indispensable and useful to move the production from Industry 4.0 to Industry 5.0.
Bari, 28 December 2021
Alessandro Massaro
About the Author
Professor Alessandro Massaro (ING/INF/01, FIS/01, FIS/03) carried out scientific research at the Polytechnic University of Marche, at CNR, and at Italian Institute of Technology (IIT) as Team Leader by activating laboratories for nanocomposite sensors for industrial robotics. He was head of the Research and Development section and scientific director of MIUR Research Institute Dyrecta Lab Srl. Actually, he carries out research activities in LUM Enterprise at LUM University -Libera Università Mediterranea(Casamassima-BA-, Italy), he is in MIUR register as scientific expert in competitive Industrial Research and Social Development, and he is currently Member of the International Scientific Committee of Measurers IMEKO and IEEE
Senior Member. He received an award from the National Council of Engineers as Best Engineer of Italy 2018 (Top Young Engineer 2018).
State of the Art
and Technology Innovation
The chapter is focused on the technological and scientific state of the art about information technology (IT) advances. Starting with Industry 4.0 enabling technologies, the scientific improvements transforming the production lines and machines in intelligent systems following the logic of Industry 5.0 are discussed. The new facilities and the new technologies are oriented on the design of flexible and dynamical production processes, taking into account the market demand which is increasingly unpredictable. Starting with the enabling technologies of Industry 4.0, the specifications of the hardware and software technologies for advances in Industry 5.0 manufacturing industries are introduced. Communication protocols able to improve sensing and actuation in production processes are also discussed. Moreover, the analysis describes the Internet of Things (IoT) protocols, IoT upgrade processes and technological improvements, where of particular interest in monitoring industrial processing is infrared thermography (IRT) for improving thermal measurements in the production environment. The chapter is also focused on the description of different levels of the company information system, where sensors monitoring production constitute the field layer. The discussion is then oriented to provide an overview about sensors communicating with the local network by protocols, and achieving intelligent and efficient sensing and actuation. All the analyzed topics are addressed for integration into an upgraded information infrastructure implementing advanced tools. The analysis is then moved to the production processes in industries by highlighting main interconnections and architectures interfacing different tools. The study also enhances the scientific approaches consolidated in Industry 4.0, by providing limits of the actual technologies and perspectives for future production upscaling. Furthermore, the chapter discusses mainly intelligent information infrastructure suitable for manufacturing industries. The chapter goal is to introduce technological elements such as artificial intelligence (AI), augmented reality (AR) and big data systems, providing knowledge gain (KG). Other important aspects are the horizontal and vertical integrations of the technologies, considering bus‐based networks and automatisms in data processing which is significant for the production advances. The chapter provides elements useful to comprehend how technologies can be implemented in flexible information architectures for innovative industrialization processes.
1.1 State
of the Art of Flexible Technologies in Industry
Industry 4.0 introduced digital technologies improving industry productivity and different facilities supporting processes. The main enabling technologies introduced by Industry 4.0 are [1–3]:
● Three‐dimensional (3D) printers connected to production software.
● AR oriented on production processes.
● Simulation tools able to optimize production processes by simulating production of different interconnected machines of different production lines.
● Horizontal integration of supply chain elements, such as human resources, supplies, products, transports, logistics, etc., and vertical integration of different production functions including product design, production processes, production quality, and end to end combination of horizontal and vertical functions.
● Cloud computing, cloud data storage, and data management in open data and big data systems.
● Cybersecurity improving security during network operations and in open systems, managing network interconnections.
These main facilities enable smart manufacturing (SM) and computer integrated manufacturing (CIM) industry processes in the fourth industrial revolution. In this scenario of enabling technologies, the information network architecture of companies plays a fundamental rule in production upgrade and in production engineering. The information digitalization is the first step for Industry 4.0 implementation, where the production machines transfer data in the local area network (LAN) and in general in the cloud environment. A particular function in Industry 4.0 improvement is the production monitoring, automated by IoT sensors [4], reading in real time the operation conditions of the whole production lines and allowing intelligent manufacturing. The control performed by sensors is more efficient for in‐line monitoring procedures, where all sensors are synchronized in order to provide the best production setting of the whole supply chain. All the phases of the supply chain are important to trace. The main parts to trace in the production processes are: (i) warehouse, (ii) production lines, and (iii) logistics. In all these parts, robotics in general improves the processes, by increasing production volumes and by assisting human work. This kind of “joint collaboration” decreases the production errors and consequently the waste materials and related costs. Other technologies such as AR [5, 6] are used for human resources training during production processing, by increasing the worker skills and supporting workers to follow interactively and continuously the production. Augmented reality aided manufacturing (ARAM) is another important topic supporting production quality [5] by means of the programming of machines, robots and production tools, by managing logistics, and by checking assembled products in the whole supply chain. AR is adopted also in manufacturing as a dynamic authoring tool monitoring simultaneously the production activities of several workstations [6], for telerobotics controlling robots from a distance, for waste reduction in production activities, for assembly support, for remote maintenance, and for computer‐aided design (CAD) applications [7]. In the Industry 4.0 scenario, AI can furthermore improve the industry production efficiency. AI algorithms are mainly indicated for machine predictive maintenance [8, 9] and for assisted production, where machine working operations are properly and automatically set in order to avoid failures [10], by decreasing or stopping machine in cases of alerting conditions. IoT sensors are very important for control and actuation thus enabling totally automated processes. A broad use of IoT sensing is related to image vision [11, 12] including IRT [13], and temperature and humidity sensors. Moreover, accelerometers provide supplementary information about anomalous vibrations indicating a possible system failure, and
1.1 Stateof theArtof leeiile Technologiees in Inddestry 3
other sensors can be applied depending on the manufacturing process to be controlled. IoT signals are processed by AI algorithms to predict the machine status in self learning modality: by analyzing historical data, the AI algorithms create the training models to test for prediction. The AI improvements represent mainly the passage from Industry 4.0 to Industry 5.0 facilities adapting automatically the production with high level efficiency, and optimizing the production processes which are previously simulated. The flexibility of the production is due to the correct choice of the sensor network architecture, of protocols and the possibility to optimize the different layers of the whole communication system of the company. A correct design of the information system allows the disposal of a modular network open to vertical and horizontal integrations introducing innovative tools and algorithms addressing the automatic production control. The layers where it is possible to operate for a flexible production are the input/output (I/O) layer, the user interface layer, the gateway layer, the IoT middleware, the processing layer, and the application layer.
1.1.1 Sensors and Actuators Layer: I/O Layer
The I/O layer is the first layer related to the production field controlled by sensors. The process of machines can be changed by actuation commands provided by the processing layer. The actuation commands must ensure the production synchronization of the whole production lines managing different production steps. In this layer, IoT devices are very important for the accuracy and reliability of the performed measurements. The data sampling is essential for a correct monitoring procedure. When the sensors control different production process steps, it is fundamental to configure and to synchronize all the sensors of the same production line. The IoT technologies are defined for the specific production process to monitor. For example, if the process is fast, it is important to select an image vision technology having a high frame rate, or sensors having a sampling time “following” the production velocity. The technologies for industrial image vision converting light into electrons are charge‐coupled device (CCD), complementary metal oxide semiconductor (CMOS), indium antimonide (InSb) infrared (IR) detectors, indium‐gallium‐arsenide (InGaAs), germanium (Ge), and mercury cadmium telluride (HgCdTe) sensors. Table 1.1 shows the working wavelengths of the IR technology.
Sensor networks are designed after an accurate analysis of the production processes, thus suggesting the correct configurations and connections of possible gateways, routers, and of device combinations. Sensor networks are implemented for point to point, star, extended star, bus, or mesh configuration. In Figure 1.1 are shown the different main network configurations. The design of the network is an important step for the realization of the correct network. The spatial allocation of the production machines and the workflow of the production define the best configuration. The network layout changes with the sensor system: the star or mesh network is typically
Table 1.1 Spectral ranges of infrared technology.
adopted for sensors, besides the bus layout is suitable for production line connections and for the information system. By considering for example a photovoltaic camp with a high number of panels, it is preferable to realize a ring type fiber optic network linking all electrical string panels. The network also assumes a hybrid configuration, especially when a new network is added and linked to an old one. Table 1.2 lists the main advantages and disadvantages of the different network layouts.
The network typology must be compatible with the network information system of the industry. In this way, the hybrid solutions potentially ensure the best network performances and flexibility. Figure 1.2a shows an example of a hybrid extended star network, constructed by merging an extended star with a mesh network, by showing an example of network reconfiguration in cases of connection failures, where the automatic principle of node commutation is managed by an intelligent algorithm detecting and predicting system failures (example of direct interaction between processing layer and production machine layers). The cases of Figure 1.2b–e are related to a possible configuration of the data transmission of a part of the network of Figure 1.2a. This example highlights the importance of adding nodes to avoid the transmission problem. The solution to add nodes to the local network must be “weighted” with the decrease of performance due to the increased complexity of the new hybrid network. The prediction of possible failures of nodes, allows to change anticipatedly a linking configuration, thus avoiding data interruptions, and preserving production control. In the prediction calculation, sensors play an important role because they detect operation conditions of production lines, status machines and product tracking. In Table 1.3 and Table 1.4 are listed the main specifications of traceability sensors able to detect the product in each production stage, and the main characteristics of transmission protocols,
Table 1.2 Advantages and disadvantages of network typologies.
Network typeAdvantages
StarThe star network manages the whole network by a single node behaving as a master node. Each node of the network can be added, removed, and reconfigured by ensuring the network operations. Network simplicity. Easy identification of errors
BusLow cost and simple layout. Connection with a simple coaxial or RJ45 cable
Disadvantages
For a failure of the central node the whole network is out of order. Bandwidth limitation
RingBidirectional transmission mode for dual ring typology (full duplex)
Tree By adding to the tree network a star and a bus layout, it is possible to allow an easy addition of nodes and a network expansion
MeshReliable and stable network type. Resistance to failure conditions also for complex layouts involving more interconnections
For a failure of the bus the whole network is out of order. Additional nodes decrease network velocity. Single direction transmission mode (half duplex)
Half duplex modality for basic ring configuration. Transmission security (if a node fails the network stops operating)
If the root node is out of order the whole network fails. Network performance decreases for complex hierarchical layouts
High time for network setting. High computational cost for complex interconnection layouts
1.1 Stateof theArtof leeiile Technologiees in Inddestry 5 respectively. Solutions for the actuation are the plug and play (P&P) solutions and programmable logic controller (PLC) hardware interfaces. For P&P systems the hardware and software components are downloaded and installed at or before run‐time. The supervisory control and data acquisition (SCADA) [30] systems are able to read production data and transmit the setpoints to the PLCs. SCADA systems typically are implemented to control system architectures by graphical user interfaces (GUIs), and behaving as a supervisor of peripheral devices such as PLC, and proportional integral derivative (PID) controllers interfacing process plant and production machinery. Typically, SCADA adopts visualization tools and synoptic graphics for real‐time data display.
Figure 1.1 Example of network configurations: (a) point to point connection; (b) bus line; (c) ring layout; (d) star connection; (e) tree layout; and (f) node meshing configuration.
Figure 1.2 Example of hybrid extended star network and failure system reconfiguration for a secure production monitoring: (a) hybrid network structure by extended star and mesh network; (b) normal configuration for data transmission to the manager central node (transmission from node 3 to node 1); (c) example of reconfiguration for an interrupted linking between node 1 (network coordinator) and node 2; (d, e) examples of reconfiguration for interrupted links between node 1 and node 2 and between node 1 and node 4 simultaneously.
The information levels of a dynamic information system, allowing the upgrade from Industry 4.0 to Industry 5.0 production, is interconnected as in Figure 1.3, where the following six main layers are distinguished:
● Sensor and actuator layer.
● Agent, firmware and user interface layer.
● Gateway layer.
● IoT middleware.
● Processing layer.
● Application layer.
Flexible technologies must act in these layers, and are fundamental to automatize all the production processes for:
● In‐line/off‐line production monitoring.
● The elimination of possible failure conditions.
● The decrease of production defects.
● The optimization of human resources and of their work.
● Business intelligence (BI) and strategic marketing.
● The optimization of the warehouse management.
● A dynamic production following the real‐time customer requests.
The main flexible technologies are integrated in robotic systems. Robots process information acquired by sensors placed inside and outside the production machines, and generating different outputs suitable for decision making, for the processing coordination, and for the system control.
The flexibility of the information system is mainly in the interconnectivity of all the layers shown in Figure 1.3, representing a standard architecture upgraded by AI and big data system working in the processing layer. Big data systems are characterized by the following features:
● Volume (dataset volumes larger than terabytes [1012 byte] and petabytes [1015 byte]).
● Velocity (velocity refers to the data generation speed).
● Variety (variety of sources with structured, semi structured and unstructured data).
● Veracity (quality of the data that is being analyzed, the non‐valuable data are classified as nose or wrong data).
Big data uses the not only SQL (NoSQL) technology. The NoSQL databases (DBs) do not use the relational model, are performed efficiently on clusters, and can be open source.
A primary important aspect concerning the production optimization is the production traceability, performed by sensors. Digital traceability is fundamental in Industry 4.0 scenarios. Automatic detection by gates installed on the production line at each production stage is able to control quality processes and production in general. Table 1.3 lists the main sensors used for product traceability.
Radio frequency identification (RFID) systems are constituted by a reader and by a TAG or transponder, enabling the electronic identification of the traced product. The active version is equipped with a lithium (Li) battery or is powered by an external source. The passive RFID is not equipped with a battery and is cheaper offering an infinite lifetime. Besides, the active TAGs are more useful when writing operations. Sensors displaced to control production transmit data using specific protocols. Table 1.4 lists some specifications of transmission protocols.
Application Layer
Production Engineering System Coordination
Big Data
Analytics
API
Processing Layer AI
IoT Middleware
Security
Gateway/ESB
Agent, Firmware, User Interface Network
Actuators Sensors
Protocols
Desi g n
D y namic Architecture
Production Machine, Robots
Figure 1.3 Layers of technologies related to an advanced technology.
Table 1.3 Main specifications of sensors used for traceability.
Sensor typeMain specifications
Barcode
QRcode
RFID
● Optical laser reading identifying only type of item
● Only read
● <20 of characters of data capacity
● Optical laser reading identifying only type of item
The agent layer interconnects the gateway for device communication with all the other components. It is based on the agent’s modular architecture suitable for robotic and production machines controlled by sensors. The IoT agent is linked to the IoT controller, and checks the framework environment during the time, by acquiring data by sensors properly programmed by firmware. Firmware is developed through software and hardware interfaces. Integrated development environment (IDE) platforms are usually adopted for firmware development as user interfaces. Agent‐based computing (ABC) tools are able to implement agent functions in modern distributed applications; multi‐agent systems (MASs) are ensembles of agents interacting in the same framework [31]. The MASs support edge and cloud computing, interconnecting innovative applications such as machine learning and blockchain.
1.1.3 Gateway and Enterprise Service Bus Layer
The gateway is an important node executing data routing and switching functions of the information network, using different protocol types. The gateway behaves as a provider, and it is interfaced between the IoT middleware and the agent layer. This layer is responsible for publishing and subscribing the services, message routing, and allows the communication between platforms. The enterprise service bus (ESB) enables service‐oriented architecture (SOA) applications. Typically, the ESB is adopted to collect the digital data sources of different technologies adopted in the same information network, such as big data systems and AI algorithm datasets, by allowing data transfer between different DB systems [32]. ESB solutions are suitable to integrate old technologies with new ones improving the information system and Industry 4.0 implementations. The gateway system covers the role of collector of the packets coming from different sensors applied in the specific production line field. Communication takes place through IEEE 802.11 (WI‐FI/5 GHz) or MODBUS standards, following the most common mechanisms for asynchronous communication including Message Queuing Telemetry Transport (MQTT) protocol [30].
The architecture of the gateway follows Industry 4.0 mechanisms, where the gateway represents the main local component made by:
● A sensor manager, able to locally manage the sensor nodes.
● A local DB manager with regulated accesses.
● An engine optimizer capable of processing advances.
● The application programming interfaces (APIs) integrating the IoT sensor network with the cloud environment by the use of mobile devices.
The IoT gateway is designed to implement the following features:
● IoT network management (local knowledge of connected devices and resources, local management of sensors or network nodes, local data availability).
● System intelligence (ability to perform process optimization locally, integrity of the information received).
● Distributed logic (the data from the nodes or sensors are stored in the local DB and managed for the central system).
The ESB is a software infrastructure carrying out services in complex SOAs and supporting horizontal, vertical and end to end integrations. The main advantage of this infrastructure is to interconnect and to interface heterogeneous technologies including big data systems, ensuring data synchronization, data security, messaging, intelligent routing managed by AI algorithms, and transformations services. The ESB assists the developer in integrating applications, therefore providing the infrastructure necessary to implement routing, translation, and other integration of features.
1.1.4 IoT Middleware
The middleware function [21] is to connect different typologies of programs. This interface is part of the architecture enabling sensor connectivity and application layers. The middleware manages important functionalities, such as collecting and selecting the received data from the IoT devices, by providing access monitoring for applications. The security is mainly performed in the middleware system by [33]:
● User identification.
● Identity management.
● A secure data communication system.
● Secure storage.
● A secure software execution environment.
● Secure contents.
● Security resistance.
The middleware improvement is related to the following specifications:
● Interoperability between devices of different technologies by managing heterogeneous interfaces.
● Managing devices performing load balancing.
● Use of API calls.
● Scalability by supporting the communication between a large number of devices.
● Big data interconnection and big data analytics tools.
● AI algorithm interconnection activating sensor data processing.
● Authentication and implementation of access control improving security and privacy.
1.1 Stateof theArtof leeiile Technologiees in Inddestry 11
● Running algorithms on different cloud services.
● Data extraction and data migration improving innovative DB and big data systems.
Middleware is classified as event based when all the components interact with each other through events, or service oriented when service providers are used for resource management in SOAs, or DB oriented, or application oriented.
1.1.5 Processing Layer
The processing layer is very important for the orientation of the industry network on facilities to Industry 5.0, including big data systems and AI algorithms, able to activate advanced data processing and the setting of all parameters of the architecture of Figure 1.3. The data processing is suitable for predictive maintenance, for machine failure predictions, for the self‐adaptive production, and in general for assisted production.
1.1.6 Application Layer
This layer is responsible for delivering various application services. These services are provided through the middleware layer. The application services are suitable for, logistics, BI, rapid prototyping, reverse engineering (RE) and other advanced industry applications.
1.1.7
File Transfer Protocols
File transfer of the company information systems is executed through file transfer protocols. The main protocols adopted by the information network systems are listed in Table 1.5.
Particularly interesting is the LoRaWAN protocol suitable for long range wide area network (WAN) wireless technology tailored for IoT interconnection, and for bidirectional communication systems. The main features of this protocol are the low power consumption, and the possibility to improve scalable wireless networks.
Table 1.5 Other protocols usable in industry.
File transfer protocols
File Transfer Protocol
Simple Mail Transfer Protocol
TCP
Details References
● Client/server
● TCP connection
● Protocol interpreter;
● Data transfer process
● Data and command separate connections [34]
● Connection oriented
● Text based
● Client/server communication [35]
● Network protocol (ISO/OSI transport level)
● Transport port (MODBUS TCP): 502, 802
● Encapsulated into Internet Protocol
● Application programming interface system call
● Security (MODBUS TCP): TLS
● Segment with a header and a data section [30] (Continued)
Table 1.5 (Continued)
File transfer protocols
MODBUS
HTTP
Hypertext Transfer Protocol
Secure
Constrained Application Protocol
Details
● Serial MODBUS RS485 (maximum 31 slaves)
● Transaction enabled by a master
● Cyclic redundancy check (CRC16) algorithm
References
● Ethernet MODBUS TCP/IP [36]
● Application protocol
● Hypertext
● Infrastructure: Ethernet, WiFi
● Model: synchronous
● Stateless protocol (communication protocol)
● Information web transfer
● Mechanism: one to one
● Network layer: IPv4 or IPv6
● Transport layer: TCP
● Transport port: 80, 443
● Standard: Internet Engineering Taskforce RFC7230
● Encoding: ASCII text
● Security: SSL or TLS [30]
● Extension of HTTP
● Secure communication
● Encryption using encrypted using TLS or SSL
● Website authentication
● Privacy protection
● Digital certificates [37]
● Internet application protocol
● Infrastructure: 6LoWPAN
● Network layer: IPv6
● Transport layer: UDP
● Transport port: 5683
● Mechanism: one to one
● Model: asynchronous
● Standard: IETF (RFC7252)
● Service layer protocol
● Wireless sensor network nodes
● Devices supporting UDP
● Security: Datagram transport layer security
● Message length: 4 byte [30, 38, 39]
Table 1.5 (Continued)
File transfer protocols
Message Queuing Telemetry Transport
Extensible Messaging and Presence Protocol
Advanced Message Queuing Protocol
LoRa
KNX
Process field net
Details
● Client/server
● Infrastructure: Ethernet, WiFi
● Messaging transport
● Model: asynchronous
● Transport port: 502, 802
● Standard: ISO/IEC, OASIS
● Publish/subscribe
● Machine‐to‐machine connectivity
● Remote connections
● Sensor communications
● TCP/IP ports: 1883, 8883
● Security: SSL, TLS
● UDP transport
● Streaming XML
● Multiple communication patterns
● Asynchronous messaging
● Publish/subscribe
● TCP transport
● HTTP transport
● Request/response
● Point to point
● Publish/subscribe
● Queuing
● Routing
● Lower physical layer
● Radio frequency (433 MHz, 868 MHz, 915 MHz)
● Geolocation capabilities
● LoRaWAN (managing data device and frequencies transmission)
● Long‐range connectivity
● Network standard
● Communications protocol
● Twisted pair bus (EHS, BatiBUS, EIB)
● Sensing and actuation
● Distributed applications
● Radio (KNX‐RF)
● Power‐line networking
● Industry technical standard for data communication
● XML
● PROFIBUS
● TCP/IP channel
References
[26, 30, 38, 39]
[30, 39]
[39]
[25, 39]
[40]
● Real‐time channel [29]
HTTP, Hypertext Transfer Protocol; KNX, Konnex; LoRa, long range; SSL, secure sockets layer; TCP, Transmission Control Protocol; TLS, transport layer security; UDP, User Datagram Protocol; XML, eXtensible Markup Language.
1.2 State
of the Art of Scientific Approaches Oriented
on Process Control and Automatisms
Technologies and architectures are fundamental for the upgrade of the company production. Different examples are provided to comprehend how innovative tools, including AI, can be applied in a new production scenario.
1.2.1 Architectures Integrating AI
The software and the hardware, in a modern information system oriented on Industry 5.0, must be integrated into a flexible information system infrastructure structured as illustrated in the model of Figure 1.4, where the first layer is represented by the hardware layer constituted by sensors and in general by IoT devices. The firmware layer is programmed in an operating system (OS) framework to provide a defined application including control and actuation processes. The AI interface is the intelligent core of the system, able to update machine setting, by considering prediction of failure conditions, and by automatically adjusting machine and robot working conditions. This automatism is a fundamental aspect of Industry 5.0 systems, where processing functions are managed in auto‐adaptive modality. The AI algorithms can be executed directly in the application layer by executing a code (python, java, visual basic, etc.), or by using objects of GUIs. In the architecture model of Figure 1.4, the processor, operates in multitasking and multiuser modality, where the central processing unit (CPU) time is divided for all the programs that work simultaneously. The AI engine and the OS manage the database management system (DBMS) able to collect digital information of the whole supply chain. Furthermore, AI is suitable to exchange data and digital commands to electronic boards, interfacing machines or robots as for intelligent actuators. Another important element for Industry 4.0 and Industry 5.0 implementations are the PLC systems enabling the actuation. The block scheme of Figure 1.5 represents a variation of the classic Von Neumann microprocessor scheme including the AI capability in the PLC system. The PLC system is constituted by:
● A random access memory (RAM).
● A microprocessor.
● Input and output (I/O) ports.
● An AI engine interfaced with a Programming Unit interface.
Figure 1.4 Hierarchical scheme of the software in Industry 5.0.
1.2.2 AI Supervised and Unsupersived Algorithms
Different technologies can be integrated to improve Industry 4.0 production processes and in general digitalization. In industrial cases, supervised and unsupervised AI image vision tools detecting defects can be adopted. In this way, image vision and 3D image reconstruction methods play an important role as for assembling processes in the tire industry [41]. The image processing of raw images requires usually a high computational cost and is applied in post‐processing modality. Concerning real‐time processing, image segmentation techniques detect most relevant defects but more information about probable defects is hidden and is not easily visible in standard industrial environments. The hidden defects are “extracted” by more accurate inspections and by the application of specific intelligent image processing algorithms enhancing anomalous features. Intelligent algorithms are usually named machine learning (ML). In Table 1.6 the main ML unsupervised and supervised algorithms are classified [42]. The supervised learning algorithm processes a known input dataset and data outputs to learn the regression/classification model. In supervised learning approaches, the training is performed by “labelled” data, selecting specific variables to focus the analysis: some data are already tagged with the requested answer, and the labeled data are adopted for the self‐learning of the algorithms predicting outcomes of the labeled variables. Unsupervised learning is the training modality of the algorithm which processes a dataset that is not classified/ labeled. In the unsupervised learning approaches the model does not need to be supervised: the models discover information and common features of the variables (attributes) and find all kinds of unknown patterns in the data. The learning phase is structured in the following sequential steps:
● Training dataset construction.
● Features vector extraction.
● Algorithm application setting data processing parameters.
● Training model construction.
Both classes of supervised and unsupervised algorithms are typically applied for data processing applications of image processing for feature classification.
A typical class of unsupervised algorithms are the clustering ones. Clustering methods are able to group objects into homogeneous classes. A cluster is a set of objects that have similarities to each other, but which have dissimilarities with objects in other clusters. The input of a clustering
Figure 1.5 Advanced PLC architecture in Industry 5.0: central processing unit (CPU), memory, I/O ports, Program Unit module, and AI upgrading industrial processes.
Table 1.6 Classification of machine learning algorithms.
Machine learning algorithm class
Continuous
Categorical
Unsupervised
● Clustering:
– K‐means
– Mean shift clustering
– Density‐based spatial clustering of applications with noise
– Expectation maximization Clustering using gaussian mixture models
– Agglomerative hierarchical clustering
● Dimensionality reduction:
– Principal component analysis
– Singular value decomposition
● Association analysis:
– Apriori
– FP‐growth
● Hidden Markov model
Supervised
● Linear regression
● Polynomial regression
● Artificial neural network
● Random forests
● Decision trees
Classification:
● k‐nearest neighbors
● Decision trees
● Logistic regression
● Naïve Bayes
● Artificial neural network
● Support vector machine
algorithm consists of elements, while the output are clusters where the elements are divided according to a similarity measure. Clustering algorithms also provide a description of the characteristics of each cluster, which is essential for decision‐making processes. Concerning the continuous class of unsupervised algorithms, the K‐means algorithm is the classical approach estimating the centroid of the clustered data named clusters. The mean shift clustering (MSC) is a sliding‐window‐based algorithm executed to find dense areas of data points defining the centroids. The density‐based spatial clustering of applications with noise (DBSCAN) has the main characteristic to find arbitrarily sized and arbitrarily shaped clusters. The expectation maximization (EM) clustering uses the gaussian mixture model (GMM) approach assuming data points distributed as a Gaussian function and characterized by mean and standard deviation. Agglomerative hierarchical clustering (AHC) group clusters follow a hierarchy represented by a tree or by a dendrogram: the root of the tree is the main cluster grouping all the samples, and the leaves are clusters with only one sample. Principal component analysis (PCA) is able to reduce the dimensionality of a dataset made up of many more or less correlated variables. The single value deposition (SVD) technique is a particular factorization of a matrix based on the use of eigenvalues and eigenvectors. Concerning the categorical class of unsupervised algorithms, the Apriori algorithm is adopted for association rules for frequent item set mining. The algorithm of FP‐growth is able to complete a set of frequent patterns by pattern fragment growth, using frequent patterns. The hidden Markov model is a statistical Markov model with hidden states. Concerning the continuous class of supervised algorithms, the regression is a statistical process that tries to establish a relationship between two or more variables. If a regression model is given an x value, this returns the corresponding y value generated by the processing of x. The linear regression differs from the classification, since the
latter is limited to discriminating the elements in a given number of classes (label), while for the linear regression approach the input is data and the system gives a real output (unlike the classification method which receives as input data and returns as output a label dataset). Polynomial regression uses the same method as linear regression, but assumes that the function that better describes the data trend is not a straight line, but a polynomial. Artificial neural networks (ANNs) are able to classify and predict data. Specifically, ANNs are made by three types of layer: the input layer, the hidden layers, and the output layer. In the input layer, the neural network receives the data in the form of inputs, activates and processes them according to the classification capacity for which it is trained, and passes the information obtained to the next layer as in neuron propagation. At each step, the starting information takes on an increasingly refined meaning due to the interpretations of the different nodes. Finally, the processed data arrive at the output layer, which collects the results. Concerning ANNs, the network is structured to learn automatically in self‐learning modality. Each perceptron has the task of categorizing objects by referring to common characteristics, following a score system calculated on each analyzed element. In AI, the perceptron represents a binary classifier selecting data input, and provides a features vector. The classifier makes its predictions based on a linear predictor function combining a set of weights with the feature vector. In AI machine learning algorithms, the perceptrons are important for their self‐learning capacity, thus addressing these tools for auto‐adaptive production solutions. A multilayered perceptron (MLP) is a particular class of feedforward ANN characterized by multiple layers of perceptrons having a threshold activation. The MLP consists of at least three layers of nodes: an input layer, a hidden layer, and an output layer. Except for the input nodes, each node is constituted by a neuron implementing an activation function. MLP utilizes a supervised learning technique called backpropagation for model training. In the class of categorical supervised algorithms, the random forest (RFo) represents a type of ensemble model, which uses bagging (the bagging aims to create a set of classifiers of equal importance) as an ensemble method and the decision tree (DT) as an individual model algorithm. Also DTs are a supervised learning tool, mainly solving classification or regression issues, capable of learning nonlinear associations and very easy to interpret and apply. DT algorithms work on both numeric and categorical data, and are categorized with respect to the output variable as categorical DT, and continuous DT. In the categorical class of a supervised algorithm, the k‐nearest neighbors (KNN) algorithm is an algorithm used for pattern recognition and for classification based on the characteristics of the objects close to the one considered. The logistic regression algorithm allows to generate a result representing a probability that a given input value belongs to a certain class. In binomial logistic regression problems, the probability that the output belongs to one class is P, while that it belongs to the other class is 1‐P (where P is a number between 0 and 1 because it expresses a probability). Naïve Bayes is a supervised learning algorithm suitable for solving binary and multi‐class classification problems ant it is based on Bayes’ theorem defining the conditional probability: let A and B be two events, and let B be a possible event having a probability of occurrence P(B) 0, if A∩B indicates the intersection of the two events (both occurred events), it is defined the conditional probability P(A|B) (probability of A conditioned by B) as:
The support vector machine (SVM) is a supervised machine learning algorithm that is used for both classification and regression purposes. For a given training dataset, the SVM training algorithm builds a model that assigns new examples to one category or the other, making it a non‐probabilistic binary linear classifier.
