A study on machine learning based modeling of skilled worker agents for production planing laerning support systems in street production

Abstract

This paper deals with machine learning based modelling of skilled worker agents for production planing learning support systems in steel production. In this paper, a try-and-error process in generating a schedule is assumed to consist of three steps: (1) select appropriate priority rules and evaluation items, (2) generate a schedule by using the priority rules, (3) evaluate of the generated schedule and revise the priority rules based on the evacuation. The scheduling generation processis modelledby using Stochastic Learning Automata, whichisakindof reinforcement learning method, to obtain thee.ective 'know-how' fora production planning learning support system. Asimulation experiment has been carried out in order to evaluate the model. The simulation results suggested that the know-hows obtained in the simulation experiments may be able to apply them into a production planning learning support system.