HELPING ENGINEERS TO ANALYSE AND INFLUENCE THE HUMAN FACTORS IN
ACCIDENTS AT WORK
2006 Institution of Chemical Engineers
Trans IChemE, Part B, May 2006
R. LARDNER_ and R. SCAIFE
The Keil Centre Ltd, Edinburgh, UK
In the
UK process industries, there are strong societal,industry and regulatory
expectations that every effort will be made to ensure the safety of process
plant, minimize injury, and protect the environment. As part of their efforts
to meet such expectations and minimize commercial loss, many companies in the
process industries have implemented an incident analysis process, which
includes some form of root cause analysis to determine the immediate and system
causes for accidents, incidents and near-misses. The existing process involved
structured evidence gathering, interviewing
by
trained staff, development of an incident time-line, identification of critical
factors, and the application of a root cause analysis model to guide
recommendations.
The
following principles were applied to the design of
the
analysis toolkit:
. tools to be based on sound analytical methods, supported by existing
research;
. methods designed to help the investigator reach their conclusions on
the basis of evidence gathered;
. methods to be suitable for use by trained investigators, who are not
human factors specialists;
. toolkit capable of being imparted via a 2-day training course,
delivered by internal company personnel;
. toolkit to permits analysis of intentional and unintentional unsafe
behaviour and identification of trends suggestive of a problem with certain
aspects of safety culture;
. provide written support, guidance and examples for investigators.
A
four-step process was developed, supported by structured worksheets, which
allowed investigators to
(1)
Accurately define and describe the behaviour(s) they wished to analyse.
(2)
Determine, on the basis of the evidence available, whether it appeared the
behaviour(s) were intentional or unintentional.
(3)
For intentional behaviour, apply ABC analysis.
(4)
For unintentional behaviour, apply human error analysis.
The
ABC model assumes the following three propositions are true:
. Behaviour is largely a function of its consequences.
. People do what they do because of what happens to them when they do
it.
. What people do (or do not do) during the working day is what is
being reinforced.
The
results of the analysis can then be turned into practical recommendations to
reduce unsafe behaviours and introduce new, safe alternatives to replace them.
Human Error Analysis
these
four stages, as the following process industry examples illustrate:
. Perception error—misperceive a reading on a display
. Memory error—forget to implement a step in a procedure
. Decision error—fail to integrate various pieces of data and
information, resulting in misdiagnosis of a process upset
. Action error—inadvertently operate the wrong device(e.g., a valve).
Trialling of Methods
Peer Review
Safety Culture Analysis
·
Visible management commitment
·
Safety communication
·
Productivity versus safety
·
Learning organization
·
Health and safety resources
·
Participation in safety
·
Risk-taking behaviour
·
Trust between management and front-line staff
·
Contractor management
·
Competence.
Conclusion
This paper describes a series
of projects in four organizations, each of whom wished to deepen their
understanding of the human factors that influence accidents and incidents at
work. Current analysis of human behaviour in incident investigation is
often relatively superficial, thus missing opportunities to improve human
performance and prevent incidents recurring. A specific weakness is
understanding of human error, which is much better understood and managed in
other domains, for example aviation.A set of human factors analysis tools were
developed such as Human error Analysis,peer review,trailling of methods and
safety culture analysis which encompassed violations, errors and aspects of
safety culture. Following a trial period and a peer review, the methods have
been implemented, and used by investigators who were typically from an
engineering background, and did not possess human factors expertise. Whilst
the results of this series of projects have been largely positive, two
challenges remain. The first is to streamline the methods to be more readily
used by busy incident investigators operating under considerable time pressure.
In doing so, a balance has to be struck between simplicity and ease-of-use, and
maintaining sufficient rigour. The second challenge is to be more
selective in the choice of delegates for this type of training. The
process and outcomes of these projects is described, with examples of how a
human factors approach can add value to existing analytical methods. Some of
the difficulties encountered are described, together with areas for future
development.