Cover Image

Artificial intelligence in manufacturing : applications and case studies /

Artificial Intelligence in Manufacturing: Applications and Case Studies provides detailed technical descriptions of emerging applications of AI in manufacturing using case studies to explain implementation.

Bibliographic Details
Corporate Author: ScienceDirect (Online service)
Other Authors: Soroush, Masoud, Braatz, Richard D.
Format: eBook
Language:English
Published: London : Academic Press, 2024.
Subjects:
Artificial intelligence > Industrial applications.
Manufacturing processes > Data processing.
Intelligence artificielle > Applications industrielles.
Fabrication > Informatique.
Electronic books.
Online Access:Connect to the full text of this electronic book
Table of Contents:
  • Front cover
  • Half title
  • Title
  • Copyright
  • Contents
  • Contributors
  • Preface
  • Chapter 1 Artificial intelligence for paints/coatings manufacturing
  • 1.1 Introduction
  • 1.2 The machine learning problem
  • 1.3 Overview of applied methods
  • 1.3.1 Traditional machine learning methods
  • 1.3.2 Deep learning methods
  • 1.3.3 CNN followed by an MLP with LAB regularization
  • 1.4 Results and discussion
  • 1.4.1 Regression
  • 1.4.2 Convolutional neural network followed by a multilayer perceptron
  • 1.4.3 CNN followed by an MLP with dropout layers
  • 1.5 Qualitative analysis of modeling approaches
  • 1.6 Experimental setup
  • 1.7 Conclusion
  • Acknowledgment
  • References
  • Chapter 2 Data-driven discovery and design of additives for controlled polymer morphology and performance
  • 2.1 Introduction
  • 2.2 Polymer crystal-growth predictions using molecular dynamics
  • 2.2.1 Measuring surface nucleation and crystal growth rates
  • 2.2.2 Characterizing effects of nucleating agents on polymer crystallization
  • 2.3 Data-driven modeling and search through a materials genome
  • 2.3.1 An evolutionary modeling approach for materials recommendation
  • 2.3.2 General perspectives on materials discovery through an evolutionary genome model
  • 2.4 Experimental validation of nucleating agents in polymers
  • 2.4.1 Experimental measurement of heterogeneous nucleation
  • 2.5 Conclusion
  • Acknowledgment
  • References
  • Chapter 3 Data, machine learning, first-principles, and hybrid models in the petrochemical industry
  • 3.1 Introduction
  • 3.2 Data type and processing in petrochemical industry
  • 3.2.1 Scalar
  • 3.2.2 One-dimensional array
  • 3.2.3 Two-dimensional matrix
  • 3.2.4 Higher-dimensional data
  • 3.2.5 Unstructured data
  • 3.2.6 Data quality control
  • 3.3 Machine learning and other data-driven models in the petrochemical industry.
  • 3.3.1 Supervised learning
  • 3.3.2 Unsupervised learning
  • 3.3.3 Reinforcement learning
  • 3.3.4 Transfer learning
  • 3.3.5 Other data-driven models
  • 3.4 First-principles models versus hybrid and data-driven models: a sliding scale
  • 3.4.1 First-principles models
  • 3.4.2 Key differences between first-principles and data-driven models
  • 3.4.3 Hybrid models
  • 3.5 Outlook
  • 3.6 Conclusion
  • References
  • Chapter 4 Perspectives on artificial intelligence for plasma-assisted manufacturing in semiconductor industry
  • 4.1 Introduction
  • 4.2 Plasma-assisted processes in semiconductor manufacturing
  • 4.3 AI for process design and production
  • 4.3.1 Reactor design optimization using surrogate models
  • 4.3.2 Generative design for optimization of parts
  • 4.3.3 Material identification and characterization
  • 4.4 AI for process optimization and efficiency enhancement
  • 4.4.1 Image processing for scanning electron microscopy/transmission electron microscopy analysis
  • 4.4.2 Process parameter space identification and inverse design
  • 4.4.3 Automated endpoint detection
  • 4.4.4 Defect classification
  • 4.4.5 Auto clean of process modules
  • 4.4.6 Virtual fabrication
  • 4.5 AI for process operation and adaptive control
  • 4.5.1 Sensor development with machine learning
  • 4.5.2 Virtual metrology and fleet matching for yield improvement
  • 4.5.3 Predictive and prescriptive maintenance
  • 4.5.4 Automated part replacement
  • 4.5.5 Predictive control and scheduling
  • 4.6 Conclusion
  • Acknowledgment
  • References
  • Chapter 5 Machine learning in reaction engineering
  • 5.1 Introduction
  • 5.2 Machine learning in catalyst design
  • 5.2.1 Dataset construction
  • 5.2.2 Representation of catalysts
  • 5.2.3 Machine learning models
  • 5.3 Machine learning for predicting reaction mechanism and kinetics.
  • 5.3.1 Mechanism prediction for gas phase reactions (including heterogeneous catalytic systems)
  • 5.3.2 Mechanism prediction for liquid/solution phase reactions (mainly organic synthesis reactions)
  • 5.4 Data-driven reaction predictions
  • 5.5 Conclusion
  • References
  • Chapter 6 Artificial intelligence in catalysis
  • 6.1 Introduction
  • 6.2 Applications of AI in improving catalytic models
  • 6.2.1 AI-enhanced microkinetic models
  • 6.2.2 AI enhanced prediction of adsorbate energies
  • 6.2.3 AI-enhanced prediction of transition state energies
  • 6.2.4 AI-enhanced sampling in ab initio molecular and metadynamics
  • 6.3 Applications of AI in computational design of catalysts
  • 6.4 AI to elucidate and design catalytic systems directly using experiments
  • 6.4.1 Structure determination from spectra
  • 6.4.2 Machine-learned models from experimental kinetics data
  • 6.4.3 AI-guided experimentation
  • 6.5 Allied and cross-cutting topics of AI in catalysis
  • 6.5.1 Uncertainty quantification
  • 6.5.2 Interpretable and domain cognizant models
  • 6.5.3 Databases for machine learning
  • 6.6 Conclusion
  • References
  • Chapter 7 Advanced manufacturing of touch-sensitive textiles
  • 7.1 Introduction
  • 7.2 Sensor interaction and machine learning
  • 7.2.1 Continuous capacitive touch localization
  • 7.2.2 Touch localization and gesture recognition
  • 7.3 Gesture recognition system
  • 7.3.1 Knitted sensor
  • 7.3.2 Measurement circuit
  • 7.3.3 Signal filtering
  • 7.3.4 Neural network architecture
  • 7.3.5 Model deployment
  • 7.4 Experiments
  • 7.5 Methods and results
  • 7.5.1 Cross-validation and evaluation results
  • 7.5.2 Resources and time performance
  • 7.6 Building interactive applications
  • 7.7 Conclusion
  • References
  • Chapter 8 Cyber-physical systems framework for AI in smart manufacturing and maintenance
  • 8.1 Introduction.
  • 8.2 Cyber-physical systems framework for maintenance and service innovation
  • 8.2.1 5C architecture for cyber-physical systems framework
  • 8.2.2 Enabling technologies for cyber-physical systems framework
  • 8.2.3 Cyber-physical systems architecture for maintenance and service innovation
  • 8.3 Development of digital twins: methodology and analytics
  • 8.3.1 Digital twin for maintenance and service innovation
  • 8.3.2 Methodology for digital twin development and deployment
  • 8.3.3 Data analytics for digital twin development and deployment
  • 8.4 Case studies
  • 8.4.1 Deep learning enhanced machine degradation assessment
  • 8.4.2 Cyber manufacturing for distributed assets
  • 8.5 Conclusion
  • References
  • Chapter 9 Dynamic data feature engineering for process operation troubleshooting
  • 9.1 Introduction
  • 9.2 Latent dynamic time-series analytics
  • 9.3 Dynamic latent variable feature extraction
  • 9.3.1 Dynamic latent feature analysis via DiCCA
  • 9.3.2 Dynamic latent relations
  • 9.4 The DELFA troubleshooting procedure
  • 9.4.1 Composite loadings for latent feature contribution analysis
  • 9.4.2 The DELFA procedure for plant-wide troubleshooting
  • 9.5 Troubleshooting plant-wide oscillations
  • 9.5.1 Preliminary analysis
  • 9.5.2 Removing the effect of TC3.OP and TI8.PV
  • 9.5.3 Analysis of the low-frequency oscillation feature
  • 9.5.4 Analysis of the high-frequency oscillation feature
  • 9.6 Comparing DiCCA with slow feature analysis
  • 9.6.1 Preliminary feature extraction based on all data
  • 9.6.2 Dynamic features and troubleshooting based on one day of data
  • 9.7 Conclusion
  • Acknowledgment
  • References
  • Chapter 10 Advanced manufacturing of biopharmaceuticals
  • 10.1 Introduction
  • 10.2 Mammalian cell bioreactor simulator
  • 10.2.1 Review of existing bioreactor models
  • 10.2.2 Foundations of bioreactor simulator software.
  • 10.3 Data-driven bioreactor process modeling
  • 10.4 Model predictive control
  • 10.4.1 Trajectory-tracking predictive control
  • 10.4.2 Critical quality attribute predictive control
  • 10.5 Data-driven modeling and closed-loop control results from mammalian cell bioreactor simulator
  • 10.5.1 Bioreactor modeling results
  • 10.5.2 Trajectory-tracking predictive control results
  • 10.5.3 Critical quality attribute predictive control results
  • 10.6 Integration of artificial intelligence in biopharmaceutical production processes
  • 10.7 Conclusion
  • Acknowledgment
  • References
  • Index
  • Back cover.