Logo image
The Elephant in the Room: Normal Performance and Accident Analysis
Conference paper   Peer reviewed

The Elephant in the Room: Normal Performance and Accident Analysis

Paul M Salmon, Natassia Goode, Erin Stevens, Guy H Walker and Neville A Stanton
Proceedings of the 17th International Conference on Human-Computer Interaction, pp.275-285
International Conference on Human-Computer Interaction (HCI International), 17th (Los Angeles, United States, 02-Aug-2015–07-Aug-2015)
Lecture Notes in Computer Science (LNCS), 9174, Springer International Publishing
2015
url
https://doi.org/10.1007/978-3-319-20373-7_26View
Published Version

Abstract

human factors ergonomics research accident causation accident prevention accident analysis methodologies
Accidents, accident causation, and accident prevention remain key themes within human factors and ergonomics research efforts worldwide. Accordingly, there are a range of well-developed models of accident causation and various methodologies to support accident analysis efforts. State of the art models propose a number of features of accident causation that go beyond operator errors and failed defenses. Once such feature now widely accepted is the notion that 'normal performance' plays a role in accidents; that is everyday behaviors not deemed to be errors or failures at the time of occurrence, are implicated in causal networks. Despite this, it is questionable whether our accident analysis methodologies are equipped to identify normal performance and its role in accidents. This paper examines this, reviewing current state of the art accident analysis methods along with their previous applications. It is concluded that, of the three methods reviewed, only one (Accimap) is currently capable of considering normal performance (at least without reclassifying it as a failure or error of some sort). The implications for accident analysis methodologies and practice are discussed and future methodological requirements are articulated.

Details

Metrics

20 File views/ downloads
722 Record Views
Logo image