| With the improvement of aircraft design technology, devices on boardhave become more and more safe. In contrast the human factors, whichcontain more variables and uncertainty, have become the main cause ofaccidents. Thus human factors are the focus and direction problem in civilaviation. Considering the increasing volume of civil aviation, studying onhuman error issues brook no delay. Maintenance has always been one ofthe main causes of aviation accident. Visual inspection is an important stepin aircraft maintenance, whose process involves a lot of human factors.Therefore, study on visual inspection cannot be ignored.More and more attention has been paid on human factors research. Itdevelops from theoretical concepts to models describing human errors,which can qualitatively analysis human factors. In unclear industry, inorder to quantitatively calculate the probability of failure, a method ofhuman reliability analysis, along with the corresponding database has beendeveloped, which deserves aviation maintenance system to learn from.Based on previous studies, this paper carried out some deeper and broadercontents. An impact factor model was set up on the basis of HumanFactors Analysis and Classification System–Maintenance. Data werecollected from experiments of inspection environmental and fatigue factorsand were used when gaining qualitative and quantitative analysis on theimpact on inspection personnel. The experiment results also can inferconclusion about strength design of composites structures. In order to savethe cost of resources, a probability of detection prediction model was buildby artificial neural network.The theoretical and experimental studies made in this study can provide qualitative and quantitative support to human factors study invisual inspection of aviation maintenance. Meanwhile the probability ofdetection curve and barely visible impact damage derived from theexperimental data can provide data support and reference for compositesstructural strength design. Subsequent studies can be made to considermore comprehensive or deeper factors, to make the analysis system furtherimproved. |