Abstract
The present study began with an assessment of the reliability and usefulness of an existing minor event coding system in a British 'high-consequence' industry. It was discovered that despite the fact that the system produced replicable data, when tested in a reliability trial the causal inferences it was producing failed to meet the normal criteria for statistical reliability. It was therefore felt necessary to create a new model of the human factors component of action in this industry, from which a model of human factors error in the same industry could be inferred. A set of codes (to facilitate statistical analysis) were deduced from this last, which were then tested in a new reliability trial. The results from this trial were very encouraging, and after a six-month pilot study in which it demonstrated its usefulness as a trend and patterning tool, the system is now being phased in within this industry.
| Original language | English |
|---|---|
| Pages (from-to) | 1-8 |
| Number of pages | 7 |
| Journal | Cognition, Technology and Work |
| Volume | 4 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - Apr 2002 |
Keywords
- error analysis
- error detection
- human factors
- near-miss reporting
- software
- risk assessment