Artificial neural networks in intelligent manufacturing

L. Monostori, D. Barschdorff

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

In intelligent manufacturing systems, unprecedented, unforeseen situations and problems are expected to be solved (within certain limits) even on the basis of incomplete and imprecise information. This goal seems to be realizable only gradually, through partial solutions integrated into today's flexible manufacturing systems. The majority of artificial intelligence (AI) applications in manufacturing rely on symbolic knowledge representation and processing. This paper draws attention to another approach, namely artificial neural networks or connectionist systems, which have the ability to integrate multiple sensor information, function in real-time, possess effective knowledge representation and can learn or adapt. For the sake of completeness a short survey of different artificial neural network structures and learning algorithms is also given, together with common applications of neural network techniques in fields different from intelligent manufacturing. The most popular back propagation learning procedure, with its most important acceleration techniques, and the competitive learning approach, which has good prospects in future applications, are highlighted.

Original languageEnglish
Pages (from-to)421-437
Number of pages17
JournalRobotics and Computer-Integrated Manufacturing
Volume9
Issue number6
Publication statusPublished - Dec 1992

Fingerprint

Artificial Neural Network
Manufacturing
Knowledge representation
Knowledge Representation
Neural networks
Competitive Learning
Flexible Manufacturing Systems
Flexible manufacturing systems
Back Propagation
Intelligent Systems
Backpropagation
Network Structure
Learning algorithms
Artificial intelligence
Learning Algorithm
Artificial Intelligence
Completeness
Integrate
Neural Networks
Real-time

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Cite this

Artificial neural networks in intelligent manufacturing. / Monostori, L.; Barschdorff, D.

In: Robotics and Computer-Integrated Manufacturing, Vol. 9, No. 6, 12.1992, p. 421-437.

Research output: Contribution to journalArticle

@article{7964697e8d81497daacb7a11d41daec5,
title = "Artificial neural networks in intelligent manufacturing",
abstract = "In intelligent manufacturing systems, unprecedented, unforeseen situations and problems are expected to be solved (within certain limits) even on the basis of incomplete and imprecise information. This goal seems to be realizable only gradually, through partial solutions integrated into today's flexible manufacturing systems. The majority of artificial intelligence (AI) applications in manufacturing rely on symbolic knowledge representation and processing. This paper draws attention to another approach, namely artificial neural networks or connectionist systems, which have the ability to integrate multiple sensor information, function in real-time, possess effective knowledge representation and can learn or adapt. For the sake of completeness a short survey of different artificial neural network structures and learning algorithms is also given, together with common applications of neural network techniques in fields different from intelligent manufacturing. The most popular back propagation learning procedure, with its most important acceleration techniques, and the competitive learning approach, which has good prospects in future applications, are highlighted.",
author = "L. Monostori and D. Barschdorff",
year = "1992",
month = "12",
language = "English",
volume = "9",
pages = "421--437",
journal = "Robotics and Computer-Integrated Manufacturing",
issn = "0736-5845",
publisher = "Elsevier Limited",
number = "6",

}

TY - JOUR

T1 - Artificial neural networks in intelligent manufacturing

AU - Monostori, L.

AU - Barschdorff, D.

PY - 1992/12

Y1 - 1992/12

N2 - In intelligent manufacturing systems, unprecedented, unforeseen situations and problems are expected to be solved (within certain limits) even on the basis of incomplete and imprecise information. This goal seems to be realizable only gradually, through partial solutions integrated into today's flexible manufacturing systems. The majority of artificial intelligence (AI) applications in manufacturing rely on symbolic knowledge representation and processing. This paper draws attention to another approach, namely artificial neural networks or connectionist systems, which have the ability to integrate multiple sensor information, function in real-time, possess effective knowledge representation and can learn or adapt. For the sake of completeness a short survey of different artificial neural network structures and learning algorithms is also given, together with common applications of neural network techniques in fields different from intelligent manufacturing. The most popular back propagation learning procedure, with its most important acceleration techniques, and the competitive learning approach, which has good prospects in future applications, are highlighted.

AB - In intelligent manufacturing systems, unprecedented, unforeseen situations and problems are expected to be solved (within certain limits) even on the basis of incomplete and imprecise information. This goal seems to be realizable only gradually, through partial solutions integrated into today's flexible manufacturing systems. The majority of artificial intelligence (AI) applications in manufacturing rely on symbolic knowledge representation and processing. This paper draws attention to another approach, namely artificial neural networks or connectionist systems, which have the ability to integrate multiple sensor information, function in real-time, possess effective knowledge representation and can learn or adapt. For the sake of completeness a short survey of different artificial neural network structures and learning algorithms is also given, together with common applications of neural network techniques in fields different from intelligent manufacturing. The most popular back propagation learning procedure, with its most important acceleration techniques, and the competitive learning approach, which has good prospects in future applications, are highlighted.

UR - http://www.scopus.com/inward/record.url?scp=0027063901&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027063901&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0027063901

VL - 9

SP - 421

EP - 437

JO - Robotics and Computer-Integrated Manufacturing

JF - Robotics and Computer-Integrated Manufacturing

SN - 0736-5845

IS - 6

ER -