Intelligent product-driven manufacturing control: A mixed genetic algorithms and machine learning approach to product intelligence synthesis

Abstract

As a specialisation of Holonic agent-based distributed manufacturing control, intelligent product-driven manufacturing control paradigm has recently emerged as one of the most promising paradigms for the development of next generation manufacturing intelligent control systems. But major issue to be solved to make this paradigm effective in real world industrial environment is related to the lack of efficiency of agent-based local decision-making approaches employed. The research work presented in this paper focuses on this pending issue and proposes and formalizes the combination of main capabilities of agent-based intelligent product-driven manufacturing control paradigm and computational intelligence genetic algorithm optimisation tool for the development of effective and efficient intelligent product driven agent-based distributed dynamic scheduling and control strategy. This challenging combination is achieved by neural network-based machine learning technique and enables enhancing manufacturing system reactivity, flexibility and fault tolerance, as well as maintaining behavioural stability and optimality.