Job Selection Method of Job Rolling Scheduling Based on Rough-Neural Network

The essential question of job rolling scheduling is how to select the job into Rolling Scheduling Window(RSW).Generally,research work always use sample heuristic method as selection rule for RSW,and most research work only consider time factor.However,multi-factors are always considered in the actual scheduling,especially priority and time.This study,therefore,propose to use rough-set to mine the selection rules for RSW from actual scheduling when the priority and time are considered synchronously,and use the rule to construct neural network to calculate which job will be selected into RSW.A rough-set method is proposed to deal with some rules conflicted to each other.