Sistema de escalonamento de processos baseado em casos

AUTOR(ES)
DATA DE PUBLICAÇÃO

1998

RESUMO

Multiple jobs that should be processed in a manufacturing plant, bank customers waiting to be served by tellers, aircrafts waiting for landing c1earances,and program tasks to be run on a computer have in common the scheduling problem that emerges whenever there is a choice concerning the order in which tasks can be performed and the assignment of tasks to servers for processing. The scheduling problem of computer processes can be automatically solved with a dedicated algorithm. One of the goals is to find a solution whenever one exists in a minimum time. The Case-Based Reasoning methodology (CBR), an approach for knowledge representation and problem solving, makes it possible to reduce processing time to get a solution when the problems become large. This work implements and evaluates a CBR-based algorithm for solving the problem of process scheduling, in critical real time systems with a monoprocessor environment. These processes must respect restrictions of ready time, computation time and deadline, besides precedence relations. The Case-Based Reasoning Real-Time Scheduler system (CBR-RTS), implemented in this work, holds, as main characteristic, the idea of simplification of a problem. This is made by substitutions of parts of the problem, that are in a case base, and adaptation of that problems aiming a solution for the presented problem. The experiments described in this work suggest that the CBR-RTS system can contribute to an expressive reduction in processing times required to schedule complex problems. In the tested examples, our algorithm performed better than the dedicated algorithm for large problems

ASSUNTO(S)

semelhanças (geometria) inteligencia artificial raciocinio analogia (logica) processamento eletronico de dados em tempo real memoria grafos de ligação

ACESSO AO ARTIGO

Documentos Relacionados