Citation

Baysari MT, McIntosh AS, Wilson JR. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 2008; 52(21): 1761-1765.

Copyright

(Copyright © 2008, Human Factors and Ergonomics Society, Publisher SAGE Publishing)

DOI

10.1177/154193120805202115

PMID

unavailable

Abstract

Nineteen Australian safety investigation reports and 13 UK safety reports were reviewed and a human error identification technique (the Technique for the Retrospective and Predictive Analysis of Cognitive Errors: TRACEr-rail version) adopted as a means of identifying and classifying errors associated with rail incidents and accidents. In the Australian sample, it was revealed that most driver errors occurred when drivers were driving or stopping their train and that these errors consistently resulted in two consequences: speeding and signals passed at danger (SPADs). In the UK, errors occurred while drivers were performing a whole range of tasks, including driving, stopping, train assembly, and failed train operations. In Australia, most 'train driving errors' were 'violations' while most 'train stopping errors' were errors of 'perception'. In the UK, a more varied range of errors was observed. Overall, the TRACEr-rail framework proved useful in categorizing driver errors from existing investigation reports and in highlighting recurring errors associated with incidents and accidents. However, the tool appeared to neglect some organizational factors contributing to driver error, and, preliminary inter-rater reliability testing revealed that some category descriptions may require further refinement. Future work will aim to develop a more Australian-friendly and reliable version of TRACEr for use in Australian rail.

Language: en