A learning-based methodology for control of scheduling in changing environment

This paper attempts the control of scheduling in the changing environment b y comparing a simple model of the life-cycle of successful human being with the control of scheduling in the industry. The four layers of this compari son are of vital importance, that represent the basic control pattern of sc heduling. Then a distributed architecture is designed based on the proposed model. The agents behavior is tested while considering one of the changing parameter i.e. machine breakdown, in a simple scheduling problem. It diffe rs from the other systems because of the distributed architecture, which em phasize the transfer of process for a particular job, during breakdown of m achine to another similar machine autonomously, having minimum job-change o ver time, by using the learning technique. The fitness of an individual is calculated based on makespan. The initial experimental results show that th e system reaches good performance both with respect to the quality of the l earned knowledge and with respect to obtaining good schedule in the changin g environment.