Using Neural Networks to predict HFACS unsafe acts from the pre-conditions of unsafe acts

Don Harris, Wen-Chin Li

Research output: Contribution to journalArticle

3 Downloads (Pure)

Abstract

Human Factors Analysis and Classification System (HFACS) is based upon Reason's organizational model of human error which suggests that there is a 'one to many' mapping of condition tokens (HFACS level 2 psychological precursors) to unsafe act tokens (HFACS level 1 error and violations). Using accident data derived from 523 military aircraft accidents, the relationship between HFACS level 2 preconditions and level 1 unsafe acts was modelled using an artificial neural network (NN). This allowed an empirical model to be developed congruent with the underlying theory of HFACS. The NN solution produced an average overall classification rate of ca. 74% for all unsafe acts from information derived from their level 2 preconditions. However, the correct classification rate was superior for decision- and skill-based errors, than for perceptual errors and violations. Practitioner Summary: A model to predict unsafe acts (HFACS level 1) from their preconditions (HFACS level 2) was developed from the analysis of 523 military aircraft accidents using an artificial NN. The results could correctly predict approximately 74% of errors.

Original languageEnglish
Pages (from-to)181-191
Number of pages11
JournalErgonomics
Volume62
Issue number2
Early online date19 Dec 2017
DOIs
Publication statusPublished - 1 Feb 2019

Fingerprint

Factor analysis
Human engineering
neural network
Statistical Factor Analysis
factor analysis
Neural networks
Aircraft accidents
Accidents
Military aircraft
accident
Aircraft
aircraft
Military
Organizational Models
organizational model
human error
Psychology

Bibliographical note

This is an Accepted Manuscript of an article published by Taylor & Francis in Ergonomics on 18 Nov 2017, available online: http://www.tandfonline.com/10.1080/00140139.2017.1407441

Keywords

  • Human Factors Analysis and Classification System (HFACS)
  • Neural Networks
  • accident analysis
  • human error
  • modelling

ASJC Scopus subject areas

  • Human Factors and Ergonomics
  • Physical Therapy, Sports Therapy and Rehabilitation

Cite this

Using Neural Networks to predict HFACS unsafe acts from the pre-conditions of unsafe acts. / Harris, Don; Li, Wen-Chin.

In: Ergonomics, Vol. 62, No. 2, 01.02.2019, p. 181-191.

Research output: Contribution to journalArticle

@article{f7ca6fdf8a704ace81ea9f3cde3efb8c,
title = "Using Neural Networks to predict HFACS unsafe acts from the pre-conditions of unsafe acts",
abstract = "Human Factors Analysis and Classification System (HFACS) is based upon Reason's organizational model of human error which suggests that there is a 'one to many' mapping of condition tokens (HFACS level 2 psychological precursors) to unsafe act tokens (HFACS level 1 error and violations). Using accident data derived from 523 military aircraft accidents, the relationship between HFACS level 2 preconditions and level 1 unsafe acts was modelled using an artificial neural network (NN). This allowed an empirical model to be developed congruent with the underlying theory of HFACS. The NN solution produced an average overall classification rate of ca. 74{\%} for all unsafe acts from information derived from their level 2 preconditions. However, the correct classification rate was superior for decision- and skill-based errors, than for perceptual errors and violations. Practitioner Summary: A model to predict unsafe acts (HFACS level 1) from their preconditions (HFACS level 2) was developed from the analysis of 523 military aircraft accidents using an artificial NN. The results could correctly predict approximately 74{\%} of errors.",
keywords = "Human Factors Analysis and Classification System (HFACS), Neural Networks, accident analysis, human error, modelling",
author = "Don Harris and Wen-Chin Li",
note = "This is an Accepted Manuscript of an article published by Taylor & Francis in Ergonomics on 18 Nov 2017, available online: http://www.tandfonline.com/10.1080/00140139.2017.1407441",
year = "2019",
month = "2",
day = "1",
doi = "10.1080/00140139.2017.1407441",
language = "English",
volume = "62",
pages = "181--191",
journal = "Ergonomics",
issn = "0014-0139",
publisher = "Taylor & Francis",
number = "2",

}

TY - JOUR

T1 - Using Neural Networks to predict HFACS unsafe acts from the pre-conditions of unsafe acts

AU - Harris, Don

AU - Li, Wen-Chin

N1 - This is an Accepted Manuscript of an article published by Taylor & Francis in Ergonomics on 18 Nov 2017, available online: http://www.tandfonline.com/10.1080/00140139.2017.1407441

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Human Factors Analysis and Classification System (HFACS) is based upon Reason's organizational model of human error which suggests that there is a 'one to many' mapping of condition tokens (HFACS level 2 psychological precursors) to unsafe act tokens (HFACS level 1 error and violations). Using accident data derived from 523 military aircraft accidents, the relationship between HFACS level 2 preconditions and level 1 unsafe acts was modelled using an artificial neural network (NN). This allowed an empirical model to be developed congruent with the underlying theory of HFACS. The NN solution produced an average overall classification rate of ca. 74% for all unsafe acts from information derived from their level 2 preconditions. However, the correct classification rate was superior for decision- and skill-based errors, than for perceptual errors and violations. Practitioner Summary: A model to predict unsafe acts (HFACS level 1) from their preconditions (HFACS level 2) was developed from the analysis of 523 military aircraft accidents using an artificial NN. The results could correctly predict approximately 74% of errors.

AB - Human Factors Analysis and Classification System (HFACS) is based upon Reason's organizational model of human error which suggests that there is a 'one to many' mapping of condition tokens (HFACS level 2 psychological precursors) to unsafe act tokens (HFACS level 1 error and violations). Using accident data derived from 523 military aircraft accidents, the relationship between HFACS level 2 preconditions and level 1 unsafe acts was modelled using an artificial neural network (NN). This allowed an empirical model to be developed congruent with the underlying theory of HFACS. The NN solution produced an average overall classification rate of ca. 74% for all unsafe acts from information derived from their level 2 preconditions. However, the correct classification rate was superior for decision- and skill-based errors, than for perceptual errors and violations. Practitioner Summary: A model to predict unsafe acts (HFACS level 1) from their preconditions (HFACS level 2) was developed from the analysis of 523 military aircraft accidents using an artificial NN. The results could correctly predict approximately 74% of errors.

KW - Human Factors Analysis and Classification System (HFACS)

KW - Neural Networks

KW - accident analysis

KW - human error

KW - modelling

UR - http://www.scopus.com/inward/record.url?scp=85038351715&partnerID=8YFLogxK

U2 - 10.1080/00140139.2017.1407441

DO - 10.1080/00140139.2017.1407441

M3 - Article

VL - 62

SP - 181

EP - 191

JO - Ergonomics

JF - Ergonomics

SN - 0014-0139

IS - 2

ER -