Knowledge Based Reactive Scheduling using Integration of Neural Nets, Simulation, Genetic Algorithms and Machine Learning
Luis C. Rabelo, Albert T. Jones
This paper addresses the development and implementation of real-time reactive scheduling and control decision-making in hierarchical manufacturing environments. The objective was to develop a prototype of a knowledge-based reactive scheduler for sequencing and multiple-machine scheduling. This prototype will serve as an important tool to study the integration of several decision-making functions and the utilization of status data to evaluate scheduling and control decision alternatives. The emphasis is on creating a predictive capability to aid in assessing the long-term system performance impact resulting from decisions made and environmental changes. This prediction capability is implemented by using neural networks, simulation, and genetic algorithms. Neural nets predict the behavior of different sequencing policies available in the system. This prediction mechanism reduces significantly the alternatives available. The contribution of the genetic algorithm to the decision-making process is the development of a new scheduling rule based on a building block procedure initiated by the neural network. The machine reaming component captures the knowledge contained in that schedule in order to avoid repetitions of the same complex process. This knowledge is in English-like statements. The research findings and the prototype developed have direct applications in the construction of real-time and reactive systems that are capable of using adaptive status data and could gracefully degrade with unforeseen situations.
Proceedings of the IFIP SIG Third Workshop KBRS '95
and Jones, A.
Knowledge Based Reactive Scheduling using Integration of Neural Nets, Simulation, Genetic Algorithms and Machine Learning, Proceedings of the IFIP SIG Third Workshop KBRS '95, Seattle, WA, USA
(Accessed February 20, 2024)