MACHINE LEARNING TECHNIQUES FOR MANAGING COMPLEXITY , CHANGES AND UNCERTAINTIES

L. Monostori
Abstract:The application of pattern recognition (PR) techniques, expert systems (ESs), artificial neural networks (ANNs), fuzzy systems (FSs) and nowadays hybrid artificial intelligence (AI) techniques in manufacturing can be regarded as consecutive elements of a process started two decades ago. On the one hand, the paper outlines the most important steps of this process and introduces some new results with special emphasis on hybrid AI and multistrategy machine learning (ML) approaches. On the other hand, agent-based (holonic) systems are highlighted as promising tools for managing complexity, changes and disturbances in production systems. Further integration of approaches is predicted. Copyright © 2002 IFAC
What problem does this paper attempt to address?