SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Calvo Olivares RD, Rivera SS, Núñez Mc Leod JE. Safety Sci. 2018; 103: 137-152.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.ssci.2017.10.023

PMID

unavailable

Abstract

Numerous theoretical models and techniques to assess human error were developed since the 60's. Most of these models were developed for the nuclear, military, and aviation sectors. These methods have the following weaknesses that limit their use in industry: the lack of analysis of underlying causal cognitive mechanisms, need of retrospective data for implementation, strong dependence on expert judgment, focus on a particular type of error, and/or analysis of operator behaviour and decision-making without considering the role of the system in such decisions. The purpose of the present research is to develop a qualitative prospective methodology that does not depend exclusively on retrospective information, that does not require expert judgment for implementation and that allows predicting potential sequences of accidents before they occur. It has been proposed for new (or existent) small and medium- scale facilities, whose processes are simple. To the best of our knowledge, a methodology that meets these requirements has not been reported in literature thus far. The methodology proposed in this study was applied to the methanol storage area of a biodiesel facility. It could predict potential sequences of accidents, through the analysis of information provided by different system devices and the study of the possible deviations of operators in decision-making. It also enabled the identification of the shortcomings in the human-machine interface and proposed an optimization of the current configuration.


Language: en

Keywords

Human error; Human reliability assessment; Accident sequences; Human-machine interface; Qualitative prospective model

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print