43
1 INTRODUCTION
Over the years, technological improvement prompted
an abatement of mishaps due to technical failures
through the use of redundancy and assurance.
However, it is difficult to discuss the reliability of a
system without considering the failure rate of every
one of its parts. One of these parts is “man”, whose
rate of failure goes to change the rate of breakdowns
of segments with which it can collaborate (Armstrong,
2001). This has featured that "human factor"
contributes significantly to the progression of
accidents, both measurably and as far as the
seriousness of the outcomes. As a result, human error
is liable for marine accidents. Over the years, many
accidents happen in marine operations due to the
human error (Islam et al., 2017a, Islam et al., 2016,
Islam et al., 2018a, Islam and Yu, 2018, Islam et al.,
2017b). In 2006, a passenger ferry Al-Salam claimed
more than 1000 lives due to human error (El-Ladan
and Turan, 2012). Moreover, in 2010 explosion of BP
Offshore oil rig in the Gulf of Mexico characterized by
fatalities and massive oil spillages. Studies of marine
accidents confirm that human failure is responsible
for about 75-96% of marine causalities globally (Wang
and Trbojevic, 2007, Williams, 1996, Youngberg and
Hatlie, 2004, Gatfield and IEng, 2006, Islam et al.,
2018a). According to the UK P&I, human failure
related accident cost maritime industry around $541m
per year (Ung et al., 2006, Islam et al., 2017a). Since
the accidents of Piper Alpha, there have been
increasing promises within the maritime community
to clearly classify and address the impact of human
failure upon maritime safety. Assessments of the role
A Review of Human Error in Marine Engine
Maintenance
R. Islam & M. Anantharaman
Australian Maritime College, University of Tasmania, Launceston, Australia
F. Khan
Memorial University of Newfoundland, St. John’s, NL, Canada
V. Garaniya
Australian Maritime College, University of Tasma
nia, Launceston, Australia
ABSTRACT: Maritime safety involves minimizing error in all aspects of the marine system. Human error has
received much importance, being responsible for about 80% of the maritime accident worldwide. Currently,
more attention has been focused to reduce human error in marine engine maintenance. On-board marine
engine maintenance activities are often complex, where seafarers conduct maintenance activities in various
marine environmental (i.e. extreme weather, ship motions, noise, and vibration) and operational (i.e. work
overload and stress) conditions. These environmental and operational conditions, in combination with generic
human error tendencies, results in innumerable forms of error. There are numerous accidents that happened
due to the human error during the maintenance activities of a marine engine. The most severe human error
results in accidents due to is a loss of life. Moreover, there are other consequences too such as delaying the
productivity of marine operations which results in the financial loss. This study reviews methods that are
currently available for identifying, reporting and managing human error in marine engine maintenance. As a
basis for this discussion, authors provide an overview of approaches for investigating human error, and a
description of marine engine maintenance activities and environmental and operational characteristics.
http://www.transnav.eu
the
International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 14
Number 1
March
2020
DOI:
10.12716/1001.14.01.04
44
of Human and Organization Errors (HOEs) is an
example of these commitments (Moore and Bea,
1995). Application of the knowledge addressing HOE
in marine and offshore operations are the collection of
human error probability data and human error
assessment and decision making (Bea, 1998, Basra and
Kirwan, 1998, Wang, 2003). Since the human factor is
one of the essential causal factors of marine mishaps,
human error probability (HEP) is a basic issue to
monitor and analyze human reliability. The aim of
Human Reliability Assessment (HRA) is to identify
the likelihood of human error for a task.
There are several methods available for HRA.
Qualitative and quantitative are two types of methods
used for the HRA to identify the human contribution
to risk. None of the available methods can be
considered better; each has strengths and weakness
and may be suitable relying upon context to be
examined, resources and accessible aptitude.
Therefore, this study focused on up to date HRA
techniques to have knowledge of the ability of the
available methods and an understanding of their
strength and weakness. Prepare a summary of the
methods. To analyze the strength and weakness the
comparison study conducted within the available
techniques. A comparison study is performed based
on the evaluation of the model, taxonomy, data, and
method that characterize each technique. Moreover,
there is potential that methods could be used out of
perspective or inappropriately, and hence it is
considered that human reliability researcher should
form a view on the ‘acceptability’ of such methods for
use in HRA and risk assessments.
2 LITERATURE SEARCH
A literature review has been conducted to ascertain
published wellsprings of information with respect to
qualitative and quantitative HRA methods, including
simulation studies and review articles. HRA studies
provided various important articles and other data
assets thorough internet search by the authors.
Moreover, a search has been conducted by the authors
for proper articles from a scope of databases, using
search terms specified by the authors. The search was
intended to collect source articles that complete HRA
techniques and ensuring validation and analysis of
journal articles. The purpose of the literature review
was to draw upon information from existing
published articles and evaluate the methodology from
source references.
2.1 Search results
The data used in this review are collected searching
through the Web of science core collection on October
9, 2017, and data-based was update in January 2019.
Data collected from the 1900-present in the field of
HRA in marine operations. There are several
keywords like “Human error”, “Human error
probability” “Human reliability assessment” “Human
Error” and search topics and “marine operations are
used. A total of 12 records were found in the database
excluding book chapters. The search results
categorized based on the temporal trend of
publications and geographical distribution. The
available methodologies for marine operations
presented in the Table 1.
Table 1. Available Methodologies for marine operations
_______________________________________________
Serial Methods
_______________________________________________
1 Cognitive Reliability and Error Analysis Method
2 Fuzzy set theory and Analytical hierarchy process
3 Human Error Assessment and Reduction
Technique by incorporating interval type-2 fuzzy
sets
4 Human Error Assessment and Reduction
Technique (HEART), Human Factors Analysis and
Classification System, Analytical hierarchy process
(AHP)
5 Hybrid approach integrating HEART with AHP
6 Cognitive Reliability and Error Analysis Method
7 Success Likelihood Index Method (SLIM)
8 Bayesian Network (BN) integrates elements from
the Technique for Retrospective and Predictive
Analysis of Cognitive Errors (TRACEr)
9 Fuzzy based Success Likelihood Index Method
10 Integration of Absolute Probability Judgment for
End points- Success Likelihood Index Method
11 Cognitive Reliability and Error Analysis Method
12 Bayesian Network (BN)
_______________________________________________
3 COMPARATIVE EVALUATION OF THE MOST
USES METHODOLOGIES
SLIM is an expert judgment method in probabilistic
reliability analysis. SLIM is a method for quantifying
the preference in a set of options. Applicability of
SLIM in assessing human reliability derives from the
consideration that human performance is affected by
different factors to assess a human response. SLIM is
a simple and flexible method based on an expert
judgment approach. The basic principle of this
method is that the likelihood of an error occurring in
a specific situation is associated with the combined
effect of PSFs (Islam et al., 2016). The SLIM procedure
is demonstrated in Figure 1.
Figure 1. Application of SLIM to estimate HEPs in
maintenance procedures of marine engine (Islam et al.,
2016)
HEART is a technique for comparing HEP and its
approach is based on the degree of error recovery.
Its fundamental basis is that in reliability and risk
equations, one is interested only in those ergonomics
factors which have a large effect on performance. The
45
factors which have a significant effect are considered
in the HEART (Islam et al., 2017a). This method is
easy to understand, fast and reliable. However, its
approach is quite subjective and heavily reliant on the
experience of the analyst (Casamirra et al., 2009). The
HEART procedure can be seen in Figure 2.
THERP is the most commonly used method in
probabilistic safety assessments (Jae et al, 1995). This
methodology includes task analyses and error
identification and representation, as well as HEPs
quantification. Probably, because of its relatively large
human error database, and its resemblance with
engineering approaches, it is used extensively in
industrial applications in comparison to other
techniques (Kirwan, 1994). THERP uses performance-
shaping factors to make judgments about specific
situations. In some cases, however, it may be difficult
to accommodate all the factors that are considered
significant. While THERP has the advantage of
simplicity, it does not account for a dependency on
human performance reliability with respect to time.
This method includes a set of tables for evaluating
HEPs that provides the basic HEP and the range of
effect factors related to the activities (Xiaoming et al.,
2005). The procedure of THERP methodology is
demonstrated in Figure 3.
Figure 3. The THERP procedure to obtain HEP (Kirwan,
1996)
Figure 2. Developed methodology for estimating the HEP for maintenance procedures of marine operations (Islam et al.,
2016)
46
Table 2. CPT for environmental factors (Islam et al., 2018)
__________________________________________________________________________________________________
Weather conditions Normal Moderate Extreme
Workplace temperature Normal Extreme Normal Extreme Normal Extreme
__________________________________________________________________________________________________
Environmental factor (poor) 0.00 0.80 0.80 0.80 0.60 1.00
Environmental factor (good) 1.00 0.20 0.20 0.20 0.40 0.00
__________________________________________________________________________________________________
BN is a probabilistic model which represents the
interaction of variables through the direct acyclic
graph and Conditional Probability Tables (CPTs). The
networks consist of nodes and edges. Each node
represents a probability of distribution either discrete
or continuous. The nodes represent a set of random
variables and edges joining the nodes represent direct
dependencies between the variables. The relationship
between the nodes is described using CPTs. All the
variables of the network are presented in a CPT. The
CPT provides a broad description of probabilistic
interaction. If there are “n”
variables
12
,, ,
n
XX X……
, in the network and
( )
i
Pa X
represents the set of parents of each
i
X
,
then joint probability distribution for the network is
estimated as:
( )
12
1
, ,...,
n
n ii
i
P X X X P X PaX
=
=
∏
(1)
where,
( )
( )
|
ii
P X Pa X
is the discrete conditional
probability distributions of
1
X
given its parents.
Thus, the following information is required to
develop a BN model.
−
12
, ,...,
n
XX X
, set of variables (nodes)
− The interaction (edges) among the variables
−
( )
( )
|
ii
P X Pa X
conditional probability distribution
for each variable
1
X
.
Table 2, represents a CPT for environmental
factors to give better understanding of a variable in
equation 1.
Bayesian Network (BN) is a mathematical graphic
based model represented by each variable as a node
with the directed links forming arcs between them.
BN provides a natural way to handle missing data,
allows a combination of data with domain
knowledge, and assists in learning about causal
relationships among variables. Moreover, BN can
provide fast responses to queries. BN has been
applied in various industries for assessing the HEP.
The procedure of BN methodology is demonstrated in
Figure 4.
In step 1, scenario selection, identification of the
maintenance activity and category of the seafarers for
the maintenance procedures of marine operations are
required.
Figure 4. Methodology developed for estimating the HEP
during the maintenance activities of marine operations
(Islam et al., 2018b, Islam and Yu, 2018)
In step 2, it is necessary to select the factors that
affect the seafarers’ error making during on-board
maintenance activities.
The final step (step 3) is to apply the developed BN
model and estimate the HEP. If there is no new
Information available regarding seafarers’
performance-affecting factor, then it will be the HEP
for that maintenance activity of marine operations.
However, if new information is available, then it is
essential to go back to the start of step 3 to add new
evidence to update the estimated HEP.
4 CONCLUSIONS
Based on the comparative analysis of the previous
section, none of the techniques have the capability of
updating probability when new information is
available except BN. Updating probability is
important to instantly reanalyze posterior HEP based
on newly available information. BN will also help
represent the relationships between human factors
and seafarers’ actions in a hierarchical structure.
Therefore, the authors suggest using BN as a
powerful technique for more accurate HEP
assessment in the maintenance activities of the marine
engine.
47
REFERENCES
ARMSTRONG, J. S. 2001. Principles of forecasting: a
handbook for researchers and practitioners, Springer
Science & Business Media.
BASRA, G. & KIRWAN, B. 1998. Collection of offshore
human error probability data. Reliability Engineering &
System Safety, 61, 77-93.
BEA, R. G. 1998. Human and organization factors:
engineering operating safety into offshore structures.
Reliability Engineering & System Safety, 61, 109-126.
EL-LADAN, S. & TURAN, O. 2012. Human reliability
analysis—Taxonomy and praxes of human entropy
boundary conditions for marine and offshore
applications. Reliability Engineering & System Safety,
98, 43-54.
GATFIELD, D. & IENG, A. 2006. Using simulation to
determine a framework for the objective assessment of
competence in maritime crisis management.
INTERNATIONAL SIMULATION AND GAMING
YEARBOOK-NEW SERIES-, 14, 44.
ISLAM, R., ABBASSI, R., GARANIYA, V. & KHAN, F.
2017a. Development of a human reliability assessment
technique for the maintenance procedures of marine and
offshore operations. Journal of Loss Prevention in the
Process Industries, 50, 416-428.
ISLAM, R., ABBASSI, R., GARANIYA, V. & KHAN, F. I.
2016. Determination of human error probabilities for the
maintenance operations of marine engines. Journal of
Ship Production and Design, 32, 226-234.
ISLAM, R., KHAN, F., ABBASSI, R. & GARANIYA, V. 2018.
Human error probability assessment during
maintenance activities of marine systems. Safety and
health at work, 9, 42-52.
ISLAM, R., KHAN, F., ABBASSI, R. & GARANIYA, V.
2018a. Human error assessment during maintenance
operations of marine systems–What are the effective
environmental factors? Safety science, 107, 85-98.
ISLAM, R., KHAN, F., ABBASSI, R. & GARANIYA, V.
2018b. Human error probability assessment during
maintenance activities of marine systems. Safety and
health at work, 9, 42-52.
ISLAM, R. & YU, H. 2018. Human Factors in Marine and
Offshore Systems. In: FAISAL KHAN, R. A. (ed.)
Methods in Chemical Process Safety. Elsevier.
ISLAM, R., YU, H., ABBASSI, R., GARANIYA, V. & KHAN,
F. 2017b. Development of a monograph for human error
likelihood assessment in marine operations. Safety
science, 91, 33-39.
MOORE, W. & BEA, R. Management of Human and
Organizational Error Throughout a Ship‟ s Life Cycle.
Proceedings of the International Conference on
Engineered Engineering Systems-Management and
Operation of Ships: Practical Techniques for Today and
Tomorrow, 1995. 172-180.
UNG, S., WILLIAMS, V., CHEN, H., BONSALL, S. &
WANG, J. 2006. Human error assessment and
management in port operations using fuzzy AHP.
Marine Technology Society Journal, 40, 73-86.
WANG, J. 2003. Technology and safety of marine systems,
Elsevier.
WANG, J. & TRBOJEVIC, V. 2007. Design for safety of
marine and offshore systems, IMarEST.
WILLIAMS, J. 1996. Assessing the Likelihood of Violation
Behaviour. University of Manchester, Manchester.
YOUNGBERG, B. J. & HATLIE, M. J. 2004. The patient
safety handbook, Jones & Bartlett Learning.
