Relativity modeling of work motivation and human error probability based on neural network

摘要

Human error behavior is jointly determined by human and environment, specially, the effect of work motivation has been emphasized to be an important factor on human performance in psychology and behavioral science. However, current Human Reliability Analysis (HRA) pay little attention to this aspect, which creates difficulties in finding the mechanisms of error behavior that could arise in the cognitive process. To fill the gap mentioned above, this paper considered work motivation factors into HRA, the hypothesis of the relationship between work motivation and Human Error Probability (HEP) in task context was given, with reference to the relationship between arousal and performance described by Yerkes-Dodson Law in psychology, then a relativity model of work motivation and HEP on different task difficulty was proposed based on neural network, and simulation experiment was carried out. The experimental results showed the work motivation-HEP relationship is of a U-shaped curve and the optimal work motivation for a difficult task is lower than an easy task.