Hugendubel.info - Die B2B Online-Buchhandlung 

Merkliste
Die Merkliste ist leer.
Bitte warten - die Druckansicht der Seite wird vorbereitet.
Der Druckdialog öffnet sich, sobald die Seite vollständig geladen wurde.
Sollte die Druckvorschau unvollständig sein, bitte schliessen und "Erneut drucken" wählen.

Artificial Intelligence-Enabled Digital Twin for Smart Manufacturing

E-BookEPUB0 - No protectionE-Book
1080 Seiten
Englisch
John Wiley & Sonserschienen am11.09.20241. Auflage
An essential book on the applications of AI and digital twin technology in the smart manufacturing sector.
In the rapidly evolving landscape of modern manufacturing, the integration of cutting-edge technologies has become imperative for businesses to remain competitive and adaptive. Among these technologies, Artificial Intelligence (AI) stands out as a transformative force, revolutionizing traditional manufacturing processes and making the way for the era of smart manufacturing. At the heart of this technological revolution lies the concept of the Digital Twin-an innovative approach that bridges the physical and digital realms of manufacturing. By creating a virtual representation of physical assets, processes, and systems, organizations can gain unprecedented insights, optimize operations, and enhance decision-making capabilities.
This timely book explores the convergence of AI and Digital Twin technologies to empower smart manufacturing initiatives. Through a comprehensive examination of principles, methodologies, and practical applications, it explains the transformative potential of AI-enabled Digital Twins across various facets of the manufacturing lifecycle. From design and prototyping to production and maintenance, AI-enabled Digital Twins offer multifaceted advantages that redefine traditional paradigms. By leveraging AI algorithms for data analysis, predictive modeling, and autonomous optimization, manufacturers can achieve unparalleled levels of efficiency, quality, and agility.
This book explains how AI enhances the capabilities of Digital Twins by creating a powerful tool that can optimize production processes, improve product quality, and streamline operations. Note that the Digital Twin in this context is a virtual representation of a physical manufacturing system, including machines, processes, and products. It continuously collects real-time data from sensors and other sources, allowing it to mirror the physical system's behavior and performance.
What sets this Digital Twin apart is the incorporation of AI algorithms and machine learning techniques that enable it to analyze and predict outcomes, recommend improvements, and autonomously make adjustments to enhance manufacturing efficiency. This book outlines essential elements, like real-time monitoring of machines, predictive analytics of machines and data, optimization of the resources, quality control of the product, resource management, decision support (timely or quickly accurate decisions).
Moreover, this book elucidates the symbiotic relationship between AI and Digital Twins, highlighting how AI augments the capabilities of Digital Twins by infusing them with intelligence, adaptability, and autonomy. Hence, this book promises to enhance competitiveness, reduce operational costs, and facilitate innovation in the manufacturing industry. By harnessing AI's capabilities in conjunction with Digital Twins, manufacturers can achieve a more agile and responsive production environment, ultimately driving the evolution of smart factories and Industry 4.0/5.0.
Audience
This book has a wide audience in computer science, artificial intelligence, and manufacturing engineering, as well as engineers in a variety of industrial manufacturing industries. It will also appeal to economists and policymakers working on the circular economy, clean tech investors, industrial decision-makers, and environmental professionals.


Amit Kumar Tyagi, PhD, is an assistant professor at the National Institute of Fashion Technology, New Delhi, India. He obtained his doctorate in 2018. He has published more than 200 papers in refereed international journals, conferences, and books, many of which are with the Wiley-Scrivener imprint. He has filed more than 25 national and international patents in deep learning, the Internet of Things, cyber-physical systems, and computer vision. His current research focuses on next-generation machine-based communications, blockchain technology, smart and secure computing, and privacy.
Shrikant Tiwari, PhD, is an associate professor in the Department of Computer Science & Engineering, School of Computing Science and Engineering, Galgotias University, Greater Noida, Uttar Pradesh India. He obtained his doctorate in 2012. He has authored or co-authored more than 75 national and international journal publications, book chapters, and conference articles. He has five patents filed to his credit. His research interests include machine learning, deep learning, computer vision, medical image analysis, pattern recognition, and biometrics.
Senthil Kumar Arumugam, PhD, is an assistant professor in the Professional Studies Department, CHRIST (Deemed to be University), Bangalore Central Campus, Bengaluru, India. He obtained his doctorate in 2014. He has received 9 awards.
Avinash Kumar Sharma, PhD, is an associate professor in the Department of Computer Science and Engineering, Sharda School of Engineering and Technology, Sharda University, Greater Noida, India. He has published about 30 research articles in national/international conferences, journals, and book chapters, edited four books and has published four patents including one design patent.mehr
Verfügbare Formate
BuchGebunden
EUR217,50
E-BookEPUB0 - No protectionE-Book
EUR187,99

Produkt

KlappentextAn essential book on the applications of AI and digital twin technology in the smart manufacturing sector.
In the rapidly evolving landscape of modern manufacturing, the integration of cutting-edge technologies has become imperative for businesses to remain competitive and adaptive. Among these technologies, Artificial Intelligence (AI) stands out as a transformative force, revolutionizing traditional manufacturing processes and making the way for the era of smart manufacturing. At the heart of this technological revolution lies the concept of the Digital Twin-an innovative approach that bridges the physical and digital realms of manufacturing. By creating a virtual representation of physical assets, processes, and systems, organizations can gain unprecedented insights, optimize operations, and enhance decision-making capabilities.
This timely book explores the convergence of AI and Digital Twin technologies to empower smart manufacturing initiatives. Through a comprehensive examination of principles, methodologies, and practical applications, it explains the transformative potential of AI-enabled Digital Twins across various facets of the manufacturing lifecycle. From design and prototyping to production and maintenance, AI-enabled Digital Twins offer multifaceted advantages that redefine traditional paradigms. By leveraging AI algorithms for data analysis, predictive modeling, and autonomous optimization, manufacturers can achieve unparalleled levels of efficiency, quality, and agility.
This book explains how AI enhances the capabilities of Digital Twins by creating a powerful tool that can optimize production processes, improve product quality, and streamline operations. Note that the Digital Twin in this context is a virtual representation of a physical manufacturing system, including machines, processes, and products. It continuously collects real-time data from sensors and other sources, allowing it to mirror the physical system's behavior and performance.
What sets this Digital Twin apart is the incorporation of AI algorithms and machine learning techniques that enable it to analyze and predict outcomes, recommend improvements, and autonomously make adjustments to enhance manufacturing efficiency. This book outlines essential elements, like real-time monitoring of machines, predictive analytics of machines and data, optimization of the resources, quality control of the product, resource management, decision support (timely or quickly accurate decisions).
Moreover, this book elucidates the symbiotic relationship between AI and Digital Twins, highlighting how AI augments the capabilities of Digital Twins by infusing them with intelligence, adaptability, and autonomy. Hence, this book promises to enhance competitiveness, reduce operational costs, and facilitate innovation in the manufacturing industry. By harnessing AI's capabilities in conjunction with Digital Twins, manufacturers can achieve a more agile and responsive production environment, ultimately driving the evolution of smart factories and Industry 4.0/5.0.
Audience
This book has a wide audience in computer science, artificial intelligence, and manufacturing engineering, as well as engineers in a variety of industrial manufacturing industries. It will also appeal to economists and policymakers working on the circular economy, clean tech investors, industrial decision-makers, and environmental professionals.


Amit Kumar Tyagi, PhD, is an assistant professor at the National Institute of Fashion Technology, New Delhi, India. He obtained his doctorate in 2018. He has published more than 200 papers in refereed international journals, conferences, and books, many of which are with the Wiley-Scrivener imprint. He has filed more than 25 national and international patents in deep learning, the Internet of Things, cyber-physical systems, and computer vision. His current research focuses on next-generation machine-based communications, blockchain technology, smart and secure computing, and privacy.
Shrikant Tiwari, PhD, is an associate professor in the Department of Computer Science & Engineering, School of Computing Science and Engineering, Galgotias University, Greater Noida, Uttar Pradesh India. He obtained his doctorate in 2012. He has authored or co-authored more than 75 national and international journal publications, book chapters, and conference articles. He has five patents filed to his credit. His research interests include machine learning, deep learning, computer vision, medical image analysis, pattern recognition, and biometrics.
Senthil Kumar Arumugam, PhD, is an assistant professor in the Professional Studies Department, CHRIST (Deemed to be University), Bangalore Central Campus, Bengaluru, India. He obtained his doctorate in 2014. He has received 9 awards.
Avinash Kumar Sharma, PhD, is an associate professor in the Department of Computer Science and Engineering, Sharda School of Engineering and Technology, Sharda University, Greater Noida, India. He has published about 30 research articles in national/international conferences, journals, and book chapters, edited four books and has published four patents including one design patent.
Details
Weitere ISBN/GTIN9781394303588
ProduktartE-Book
EinbandartE-Book
FormatEPUB
Format Hinweis0 - No protection
FormatFormat mit automatischem Seitenumbruch (reflowable)
Erscheinungsjahr2024
Erscheinungsdatum11.09.2024
Auflage1. Auflage
Seiten1080 Seiten
SpracheEnglisch
Dateigrösse34519 Kbytes
Artikel-Nr.17531258
Rubriken
Genre9201

Inhalt/Kritik

Leseprobe

1
Machine Learning Fundamentals

Renugadevi A. S.1*, R. Jayavadivel2, Charanya J.1, Kaviya P.1 and Guhan R.1

1Department of Electronics and Communication Engineering, Kongu Engineering College, Erode, India

2Department of CSE, Alliance College of Engineering and Design, Alliance University, Karnataka, India
Abstract

Machine learning (ML) is a topic of study focused on comprehending and developing learning methods, or methods that use data to enhance performance on a certain set of tasks. It is considered to be a component of artificial intelligence. The Types of learning in Machine Learning are Supervised Learning: uses labeled data for model training, Unsupervised Learning: uses unlabeled data for model training. When labeled data is not available (there is no result to predict), the learning purpose is to find hidden similarities, groups or clusters among examples, or to determine characteristics in the data structure. Reinforcement Learning: consists of a trained agent that learns on the basis of rewards or penalties. The Model techniques used in machine learning based models are: 1) Classification: prediction task of categorical values in supervised learning. 2) Regression: prediction task of continuous values in supervised learning. 3) Clustering: find groups or similarities in data in unsupervised learning. 4) Dimensionality reduction (DR): reduce the number of variables/features in data in unsupervised learning. Among the types of learning, each machine learning consists of variety of algorithms and performance measures, which is aligned with various model techniques. This chapter focuses on all the types of machine learning algorithms such as Support vector machine, Discriminant Analysis, Naïve Bayes, K nearest neighbor, K Means, Decision tree, principal component analysis, etc.

Keywords: Machine learning, supervised learning, unsupervised learning, reinforcement learning
1.1 Introduction

Computational models known as machine learning algorithms allow computers to learn from data and make judgments or predictions without the need for explicit programming. They may be divided into three primary groups: clustering, regression, and classification. Classification algorithms, like Decision Trees and Support Vector Machines, give labels or categories to incoming data. Regression methods, such as Polynomial and Linear regression, forecast continuous numerical values. Clustering methods such as K-Means and Hierarchical Clustering combine related data points without the need for predetermined labels. These algorithms are essential for a wide range of applications, such as natural language processing and picture identification, since they can adjust to the specific issue and goal at hand. New methods keep coming up as the area develops, which carries on the machine learning progress.
1.2 Classification

Classification algorithms are used to classify test data based on prior learning. Among them, the model learns from previous data and separates the test data into different groups.

Types of Classifiers

Binary classifier: A classifier whose results have exactly two categories is called a binary classifier [1]. Example: Normal or abnormal, yes or no

Multi-class classifier: When the results of the classifier have more than two classes, it is called a binary classifier based on multiple classifiers. Example: Different stages of skin cancer, different types of products

The classification algorithm can be further divided as shown in Figure 1.1.
1.2.1 Linear Model
1.2.1.1 Logistic Regression
The probability of test items is predicted using supervised machine learning techniques like logistic regression. A binary classifier is what logistic regression is. The data is represented as 1 or 0 in the binary output variable. However, the resulting value is between 0 and 1. The S-shaped sigmoid function is used here. Linear regression models theoretically predict values as a function of X. The general linear regression model has a simple equation expressed as shown in below Figure 1.2.

Figure 1.1 Classification of test data using algorithm.

Figure 1.2 Data representation in logistic regression.
1.2.1.2 Support Vector Machine
SVM is a supervised learning system that can transform input data into a higher-order space. Data classification is done using a hyperplane with maximum. SVM can handle large amounts of data and is a widely used tool in machine learning for binary classification problems. Classification methods usually involve a set of training-test datasets.

There are many specifics of the training process and cost estimates for each event. The purpose of SVM is to obtain a model that predicts the objective value of the given data based on training data as shown in Figure 1.3. This function converts the training vectors to higher order. SVM finds the maximum marginal linear separating hyperplane in highdimensional space. If more than one feature is available, you need to select part of the input before moving to SVM.
1.2.2 Nonlinear Model
1.2.2.1 K-Nearest Neighbor
A supervised machine learning technique called K-Nearest Neighbor is used to categorize an example within a set of kNNs shown clearly in Figure 1.4. At first, the K value corresponds to a low value. Sort the test data into groups based on increased similarity by comparing how similar the test data is to each category. Regression and classification issues can benefit from its application. KNN does not learn by training, which is why it is referred to as a lazy learner.

Figure 1.3 SVM system.

Figure 1.4 Categorization of set using K-nearest neighbor.

The Euclidean distance between two points is [2]:
1.2.2.2 Naive Bayes
Naive Bayes is a hedonic method based on the Bayesian manager making predictions and assigning data x to list i with the largest return probability P. It specifically reduces the overhead of including each feature in the prediction class that contains it. Naive Bayes shows a competitive advantage over more traditional and advanced classification methods such as decision trees and neural networks. Due to its short learning time, it is also a good classifier that can easily process high-dimensional data. [3] Bayesian classification is about tracking learning and actual grouping strategies. It takes simple prediction models and allows us to express model uncertainty in a way that calculates the consequences of the event. It can solve many diagnostic and prognostic problems. Bayesian inference is used as a learning method that appears in Naive Bayes text.

A simple statement of Bayes´ theorem is as follows:
1.2.2.3 Decision Tree
The decision tree classifier provides a clear classification model that is valid in many applications.

The nodes of the tree represent tests for a particular item and branches in different directions depending on the value shown in Figure 1.5. A page represents one category of notes. According to the test results, the lines of the educational system are separated by transferring data from the roots to the leaves.

Figure 1.5 Classification of decision rule.
1.2.2.4 Random Forest
A supervised machine learning approach called Random Forest is used to address regression and classification issues. It´s an ensemble learning technique that uses many classifiers to tackle challenging issues. To boost accuracy, test data is classified using numerous decision trees in random forests, and then the data is averaged as shown in Figure 1.6. Accuracy will increase with more wood utilized, but overfitting will become an issue.
1.3 Regression

Data scientists can use regression to predict a continuous variable (y) using the mathematical standard for the outcomes of one or more predictions (x). Linear regression is probably the most commonly used form of regression analysis given its ease of estimation and forecasting [4].

Figure 1.6 Classification of data set.
1.3.1 Linear Regression

One of the most popular and simple machine learning methods used for predictive analytics is linear regression. Predictive regression describes what is predicted, while linear regression estimates constants such as age, salary, and other variables. The variance of the variable (y) varies as a function of the independent variable; this shows the relationship between the variable and the variable (x). A regression line is a line that attempts to fit data between the variables and the independent variables.

The equation of the inverted line is y = a0 + a * x + b.

Here,
y is a variable.
x = foreign variable
a0 is the intersection point of the line.

The following highlights other differences between the two forms of linear regression: Basic Linear Regression An independent variable is utilized in simple linear regression to forecast the value of the dependent variable.
1.3.2 Multiple Linear Regression

Using many independent variables, this technique forecasts a variable´s value as shown Figure 1.7.

Several uses for linear regression include:
Sales forecasting and model...
mehr

Autor

Amit Kumar Tyagi, PhD, is an assistant professor at the National Institute of Fashion Technology, New Delhi, India. He obtained his doctorate in 2018. He has published more than 200 papers in refereed international journals, conferences, and books, many of which are with the Wiley-Scrivener imprint. He has filed more than 25 national and international patents in deep learning, the Internet of Things, cyber-physical systems, and computer vision. His current research focuses on next-generation machine-based communications, blockchain technology, smart and secure computing, and privacy.

Shrikant Tiwari, PhD, is an associate professor in the Department of Computer Science & Engineering, School of Computing Science and Engineering, Galgotias University, Greater Noida, Uttar Pradesh India. He obtained his doctorate in 2012. He has authored or co-authored more than 75 national and international journal publications, book chapters, and conference articles. He has five patents filed to his credit. His research interests include machine learning, deep learning, computer vision, medical image analysis, pattern recognition, and biometrics.

Senthil Kumar Arumugam, PhD, is an assistant professor in the Professional Studies Department, CHRIST (Deemed to be University), Bangalore Central Campus, Bengaluru, India. He obtained his doctorate in 2014. He has received 9 awards.

Avinash Kumar Sharma, PhD, is an associate professor in the Department of Computer Science and Engineering, Sharda School of Engineering and Technology, Sharda University, Greater Noida, India. He has published about 30 research articles in national/international conferences, journals, and book chapters, edited four books and has published four patents including one design patent.
Weitere Artikel von
Arumugam, Senthil Kumar
Hrsg.