Human Reliability Analysis of Special Operation Based on RBF Neural Network
Aiming at the analysis to the randomness, fuzziness and uncertainty of human error, the nonlinear dynamics and fault tolerance processes of human error are put forward based on the radial basis function (RBF) neural networks. Taking the crane operator operation as the analysis example, firstly, the indexes system about the human reliability prediction is constructed based on the factors of the operator, the communion interface, the operating circumstance, characteristic and organization, and the indexes are quantified. Secondly, according to human reliability analysis (HRA) theory and the scene record, the human error data are counted, and the human error rates are given. Finally, with the analysis to operators tiredness and emotion, information channels, operation complexity and time margin, lighting and wind power conditions, working pressure and safety supervision, the crane operation human reliability model is set up based on the RBF neural network. The results indicate that the RBF prediction include the operation reliability as well as the cognitive reliability. So it needs no HRA events tree and easily operate; the predictions are much match with the observations, the error is only 0.03percent
Wang Hongde Ma Chengzheng
School of Civil and Safety Engineering Dalian Jiaotong University Dalian, China Department of Transportation & Economic Management Liuzhou Railway Vocational Technical College Liuz
国际会议
北京
英文
564-567
2011-09-23（万方平台首次上网日期，不代表论文的发表时间）