A Novel Human Reliability Prediction Method Based on GOMS

Human reliability is a primary problem for flight safety. The paper proposes a new human reliability prediction method, addressing the detailed modeling of cognitive task processes. This human reliability method built based on the CREAM (Cognitive Reliability and Error Analysis Method) adapts the cognitive task analysis from GOMS model to establish a complete task analysis process and optimizes the prediction process of human reliability. This method can output detailed reliability result step by step with task analysis method of GOMS. A man-machine cooperation experiment has been executed and the results verify the usability and correctness of the model. The results show that the HRR-GOMS model can predict human reliability more accurately, and the model can output behavioral factors based on cognitive task analysis. Therefore, the analysis results by this novel method can guide the design of man-machine interface and pilot training more targeted.