RULE-BASED STUDY ON THE FAULT DIAGNOSIS ALGORITHM OF EMERGENCY RESPONSE PROCESS
Even though the emergency is unforeseeable and cannot avoid sometimes, the rapid and effective emergency response is a very important way to reduce the casualty loss.When the disasters happen, the wrong decisions and redundant flows will only make the disaster even heavier.So it is very important to check out the fault flows in response processes, and provide a better process.This research proposes a rule-based algorithm on the emergency response process fault diagnosis.Two main reasons causing faults are summarized.A fault diagnosis model has been presented.Uncertainty knowledge representation, which is based on the production rule, is used to build the rule base.And we employ BP neural network to get the confidence, and genetic algorithm to diagnose faults.The model is validated by an earthquake case, and the result has shown that the developed model can successfully diagnose the existing processes of emergency response model.Besides, a more reasonable process is presented, which will be a very valuable suggestion for decision-makers.
Emergency Management Process Management Fault Diagnosis Neural Networks Genetic Algorithm
Siqing Shan Jihong Shi Jie Ren Zhongjun Hu
Department of Information System, School of Economics and Management, Beihang University, Beijing 100191, China
国际会议
The 12th International Conference on Industrial Management(第十二届工业管理国际会议)
成都
英文
15-22
2014-09-03（万方平台首次上网日期，不代表论文的发表时间）