Seafarers International Research Centre (SIRC)

The causes of maritime accidents in

the period 2002-2016

Acejo, I., Sampson, H., Turgo, N., Ellis, N., Tang, L.

Nov 2018
The causes of maritime accidents in the period
2002-2016

Acejo, I., Sampson, H., Turgo,N., Ellis, N., Tang, L.

Seafarers International Research Centre (SIRC)

Cardiff University
52 Park Place
Cardiff CF10 3AT

Tel: +44 (0)29 2087 4620

[email protected]
www.sirc.cf.ac.uk

November 2018

ISBN: 1-900174-51-0
Contents

Introduction .................................................................................................................................................. 1
Table 1: Numbers of accident investigation reports by country of origin............................................ 1
Method of Analysis ....................................................................................................................................... 2
Accident Types.............................................................................................................................................. 2
Table 2: Types of accident .................................................................................................................... 2
Immediate and contributory causes of different types of accidents ................................................... 3
Collision, close quarters and contact accidents ................................................................................... 3
Figure 1: Immediate causes of collision, close quarters and contact accidents.................................. 4
Figure 2: Contributory causes of collision, close quarters and contact accidents ............................... 5
Grounding ............................................................................................................................................. 5
Figure 3: Immediate causes of grounding .......................................................................................... 6
Figure 4: Contributory causes of grounding ......................................................................................... 7
Fire and explosion................................................................................................................................. 7
Figure 5: Immediate causes of fire and explosion ................................................................................ 8
Figure 6: Contributory causes of fire and explosion............................................................................. 8
Lifeboat accidents................................................................................................................................. 9
Figure 7: Contributory causes of lifeboat accidents ........................................................................... 10
Other accidents .................................................................................................................................. 10
Figure 8: Immediate causes of other accidents.................................................................................. 11
Figure 9: Contributory causes of other accidents .............................................................................. 11
Immediate causes of all types of accidents ........................................................................................ 12
Figure 10: Immediate cases (all accident types)................................................................................. 12
Contributory causes of all types of accidents..................................................................................... 12
Figure 11: Contributory causes (all accident types) ........................................................................... 13
Immediate and contributory causes of all types of accidents combined........................................... 13
Table 3: Overall causes of all types of accident ................................................................................ 14
Conclusion .................................................................................................................................................. 14
References .................................................................................................................................................. 15
Introduction
Maritime safety has undoubtedly improved in the last century as a result of a combination of factors
including: technological advancement; better training; and regulatory development (Allianz, 2012).
However, despite such improvement seafaring remains a relatively dangerous occupation (Hansen, 1996;
Roberts and Marlow, 2005; Borch et al., 2012). One way to improve safety at sea is to ‘learn’ from past
accidents. For this purpose, maritime authorities around the world invest a considerable amount of
resource in investigating accidents and producing reports. Most reports offer a detailed account of what
took place and attempt to identify all the relevant factors and contributory causes. While they frequently
provide rich information, meticulous analysis and detailed insight, such accident reports are generally
read as isolated documents and therefore fail to shed light on general patterns or trends. To identify
patterns, and lessons from accidents, it is helpful therefore to consider such documents ‘en masse’ and
to systematically aggregate their findings as far as is reasonable. This is the aim of this report which builds
upon an earlier paper which was published as part of the SIRC symposium proceedings in 2013 (Tang et
al, 2013).

The report constitutes an analysis of accident investigation reports that have been published on line by
the (UK) Marine Accident Investigation Branch (MAIB), the Australian Transport Safety Bureau (ATSB), the
(US) National Transportation Safety Board (NTSB), the Federal Bureau of Maritime Casualty Investigation
in Germany, and the Danish Maritime Accident Investigation Board (DMAIB) in the period 2002-2016
(inclusive). Accident investigation reports from Maritime New Zealand have only been included for the
period 2002-4. This is because their publication was discontinued in 2004.

In preparing this report, a total of 693 accident reports were analysed. An interim paper was published in
2013 outlining the results from the first phase of the study which included 319 accident reports from the
period 2002-2011. We subsequently analysed a further 374 reports in Phase 2 which were published in
the period 2012-2016 (inclusive). Table 1 gives the breakdown of the accident reports by country of origin.

Table 1: Numbers of accident investigation reports by country of origin

Country of origin Frequency %

United Kingdom 203 29.3
Australia 145 20.9
U.S.A. 57 8.2
New Zealand 43 6.2
Germany 137 19.8
Denmark 108 15.6
Total 693 100.0

1
Method of Analysis

The reports which were analysed were published online in the period of the study. No sampling was
applied. Each report was given a first reading by two different members of the research team (five
researchers undertook the analysis in 2002-2011, and two analysed the reports for the period 2012-2016).
In the first phase of the research the team was split into two pairs and then cases which caused some
concern with regard to classification were discussed with a moderator (the fifth member of the team).
The categorisation which was developed in the first phase of the research was then applied consistently
throughout the whole period of the analysis. The accident causes were further assigned a status as either
the ‘immediate cause’ or a ‘contributory cause’ of the accident concerned. Immediate causes refer to
those which directly lead to the accidents (at the end of error chains) while contributory causes (which
could be multiple) are defined as those that either lead to the immediate cause of an accident, or create
the conditions in which the immediate/contributory causes are likely to arise. The respective pairs of
researchers individually read accident reports and categorised causation alongside the assignation of
causes as ‘immediate’ or ‘contributory’. Having read each report the pair members consulted with each
other to check on consistency. Where there were discrepant analyses the pair initially re-read the reports
concerned and when disagreement continued or they were uncertain of their interpretation they referred
the case to the moderator. The results for these reports were then combined using Microsoft Excel.

Accident Types

The most common kind of accident was identified as ‘collision, close quarters and contact’ (35.8%). This
was followed by grounding which constituted 17% of the cases. About 9.8% of cases were associated with
fire and explosion, 3.3% of cases were related to lifeboats and 34.2% were described by investigators as
being other kinds of accidents. These disparate events included, for example, crane failure, man
overboard [sic], cargo loss, engine room flooding, trip and fall, parting of mooring lines, and oil spills.

Table 2: Types of accident

Types of Accident Frequency %

Collision, close quarters & contact 248 35.8
Grounding 118 17.0
Fire and explosion 66 9.8
Lifeboat 23 3.3
Other 238 34.2
Total 693 100.0

2
Immediate and contributory causes of different types of accidents

Collision, close quarters and contact accidents

The most common immediate cause of collision, close quarters and contact accidents was found to be
the maintenance of an ‘inadequate lookout’ which was identified in 24.6% of cases.

Inadequate lookout

[…] the second officer had been in charge of only 10 bridge watches. Therefore,
he had not been tested in a variety of shipping situations. As the master had
only known the second officer for about 2 weeks, it is astonishing that he was
sufficiently confident of the OOW’s abilities to entrust him with the bridge
watch in the Dover Strait, one of the busiest shipping lanes in the world, at
night and without a lookout for support.

Collision between general cargo vessel Paula C and bulk carrier Darya Gayatri

‘Failure in communication’ was identified in 15.3% of cases, ‘poor judgment’ (14.1%), and ‘pilot
error/mishandling’ (12.9%) also featured as prominent immediate causes of collision, close quarters and
contact accidents.

In terms of contributory causes, ‘ineffective use of technology’ was most commonly identified by
investigators. This was found to be a contributory cause in 24.2% of cases. Ineffective use of technology
was identified as a cause/contributory cause of an accident when investigators found that technology had
not been used to its full potential. An additional category of ‘inappropriate use of technology’ was applied
to misuse of technology and this comprised 6.9% of contributory causes. When ineffective and
inappropriate use of technology were aggregated we found that technology was reported to have
contributed to accidents in more than a third of the cases (31%).

Inappropriate use of technology

The recorded AIS data did not permit accurate determination of the scene of
collision. There was evidently a GPS error. The officer in charge of the
navigational watch on each vessel failed to verify the GPS positions displayed
with another system, such as radar, or visual bearings.

Collision between MV Francisca and MV RMS Bremen

3
‘Failure in communication’ was also identified as a major contributory cause and this was identified in
20.6% of cases. ‘Weather/other environmental factors’ (19.8%), ‘third party deficiency’ (14.1%),
‘inadequate lookout’ (12.5%), ‘inadequate risk management’ (11.7%) and ‘distraction’ (10.9%) were also
identified by investigators as contributory causes of collision, close quarters and contact accidents.

Figure 1: Immediate causes of collision, close quarters and contact accidents

30.0
24.6
25.0

20.0
15.3
14.1
15.0 12.9

10.0 7.3 6.5

5.6
5.0 3.6 3.6 3.2 2.8 2.8
2.4 2.0 1.6 1.6
1.2 0.8 0.8 0.4 0.4
0.0

4
Figure 2: Contributory causes of collision, close quarters and contact accidents

30.0
24.2
25.0
20.6 19.8
20.0
14.1
15.0 12.5 11.7
10.9
9.7 9.7 9.7 9.3
10.0 8.1
6.9 6.9 6.0 6.0
5.6
4.0 3.6
5.0
1.2 0.8 0.4
0.0

Grounding

In terms of grounding incidents, ‘failure in communication/coordination’ was found to be an immediate

cause in 16.1% of cases. It was also identified as a contributory cause in 24.6% of cases. Other immediate
causes of grounding were found to be: ‘inadequate lookout’ (11.9%); ‘poor judgment’ (11%); ‘fatigue’
(9.3%); and ‘technical failure’ (8.5%). Third party causes of grounding such as ‘pilot error/mishandling’
were identified in 6.8% of cases.

‘Ineffective use of technology’ (31.4%) and ‘inadequate risk management’ (28.8%) were most frequently
identified as contributory causes in cases of grounding. ‘Inappropriate use of technology’ was identified
as a contributory cause in 7.6% of cases and ‘ineffective use of technology’ was reported by investigators
to have contributed to 31.4% of groundings amounting to a total of 39% of cases in which technology
played a contributory role.

Inadequate risk management was also commonly identified as a contributory cause and was found to
play a part in 28.8% of cases.

5
 Inadequate risk management

[…] there had been insufficient passage planning for the voyage; in
particular, for the transit through the Little Russel, the extremely low tide
and effect of squat were not properly considered. This resulted in the bridge
team being unaware of the limits of safe water available and thus, despite
their good positional awareness, they headed into danger without
appreciation of the risk.

Grounding and flooding of ro-ro ferry Commodore Clipper

The next most common contributory causes of grounding were ‘failure in communication’ (24.6%) and
‘third party deficiency’ (17.8%). Fatigue’ (12.7%), ‘weather/other environmental factors’ (12.7%) and ‘rule
violation’ (12.7%) were also identified as common contributory factors in grounding reports.

Figure 3: Immediate causes of grounding

18.0
16.1
16.0
14.0
11.9
12.0 11.0

10.0 9.3
8.5
8.0 6.8 6.8 6.8
5.9 5.9 5.9
6.0
4.2
4.0 2.5
1.7 1.7
2.0 0.8 0.8 0.8 0.8
0.0

6
Figure 4: Contributory causes of grounding

35.0
31.4
28.8
30.0
24.6
25.0

20.0 17.8

15.0 12.7 12.7 12.7

9.3 9.3 9.3
10.0 7.6 7.6
5.9
4.2 4.2 4.2 4.2 3.4
5.0 2.5 2.5
0.8
0.0

Fire and explosion

‘Inadequate risk management’ prominently featured both as an immediate (27.3%) and contributory
(43.9%) cause of fire/explosion. Other immediate causes such as ‘third party deficiency’ (18.2%),
‘technical failure’ (18.2%), and ‘inappropriate/ineffective maintenance (16.7%) were also commonly
found by investigators to have resulted in fire/explosion.

Technical Failure

The water mist system should have started automatically but because of
inappropriate placement of smoke and heat detectors it did not. Furthermore it
could not be engaged immediately by using the control panel by the engine
room entrance. It has not been possible to establish why the system did not
start when the CE pushed the buttons on the panel.

Engine room fire Maersk Master

7
In terms of contributory causes, ‘poor emergency response’ was identified as a factor in more than a
quarter of cases (27.3%). ‘third party deficiency’ (18.2%) and ‘inappropriate/ineffective maintenance’
(16.7%) were also identified by investigators as important contributory causes of fires and explosions.
Both ‘inadequate training/experience’ and ‘poor design’ were identified as immediate and contributory
causes of fire and explosion accidents.

Figure 5: Immediate causes of fire and explosion

30.0 27.3

25.0

20.0 18.2 18.2

16.7
15.0

10.0
6.1
4.5
5.0 3.0
1.5 1.5 1.5 1.5
0.0

Figure 6: Contributory causes of fire and explosion

50.0
43.9
45.0
40.0
35.0
30.0 27.3
25.0
18.2 16.7
20.0
15.0 10.6 10.6 9.1
10.0 7.6 6.1 6.1
3.0 1.5 1.5 1.5
5.0
0.0

8
Lifeboat accidents

In relation to accidents associated with lifeboats investigators most commonly identified the immediate
causes as ‘inappropriate/ineffective maintenance’ (26.1%) and ‘inadequate training/experience’ (21.7%).
‘poor design’ and ‘poor judgment’ were identified as immediate causes in 17.4% and 13% of lifeboat
accidents respectively.

While ‘inadequate risk management’ was identified in 8.7% of cases as an immediate cause of lifeboat
accidents, it was more frequently regarded as a contributory cause and was cited as such in 60.9% of
cases. Inadequate training/experience (30.4%), third party deficiency (21.7%), and lack of manufacturers’
guidance (21.7%) were also frequently identified as contributing to lifeboat accidents. Poor design
(17.4%), inappropriate/ineffective maintenance (17.4%), failure in communication (8.7%), rule violation
(8.7%), and technical failure (8.7%) were also identified as significant contributory causes.

Figure 6: Immediate causes of lifeboat accidents

30.0
26.1
25.0
21.7
20.0 17.4

15.0 13.0

10.0 8.7 8.7

4.3 4.3
5.0

0.0

9
Figure 7: Contributory causes of lifeboat accidents

70.0
60.9
60.0

50.0

40.0
30.4
30.0
21.7 21.7
20.0 17.4 17.4
8.7 8.7 8.7
10.0 4.3 4.3
0.0

Other accidents

The category ‘other accidents’ consisted of all other accidents deemed worthy of investigation but not
related to collision, grounding, fire and explosion and lifeboats. For example accidents associated with
crane operations or resulting in oil spills were counted as ‘other accidents’. In these kinds of accidents,
‘inadequate risk management’ was frequently identified as both as an immediate and contributory cause
(in 34.9 and 34.5% of cases, respectively). Other immediate causes included ‘third party deficiency’
(12.2%) and ‘poor judgment’ (11.3%) alongside inappropriate/ineffective maintenance’ ‘poor design’
‘technical failure’ and ‘weather/other environmental factors’.

Weather/other environmental factors

The two crew members had proceeded onto the aft mooring deck (Figure 1) to
secure a coiled mooring rope that had loosened in its stowed position. They
were struck by a large wave, which washed them overboard, causing their
respective lifelines, which were secured to the vessel, to part.

Persons washed overboard during heavy weather with loss of two lives Timberland

10
Third party deficiencies featured as a contributory cause in 17.6% of cases. Weather/other environmental
factors (16%), inadequate training/experience (12.6%), poor judgment (10.1%), and
inappropriate/ineffective maintenance (9.7%), were also regarded as contributory causes of ‘other
accidents’.

Figure 8: Immediate causes of other accidents

40.0
34.9
35.0
30.0
25.0
20.0
15.0 12.2 11.3
8.8 8.0
10.0 7.1 6.7
5.5
5.0 1.7 1.7 1.7 1.7 1.7 0.8 0.8 0.8 0.4 0.4
0.0

Figure 9: Contributory causes of other accidents

40.0
34.5
35.0
30.0
25.0
20.0 17.6
16.0
15.0 12.6
10.1 9.7
10.0 8.0 8.0 7.1
5.5 5.0 5.0
5.0 3.4 2.5 2.1 2.1
1.7 0.8 0.4 0.4 0.4 0.4 0.4
0.0

11
Immediate causes of all types of accidents

For all accident types, the immediate causes most frequently identified by investigators were as follows:
inadequate risk management (17%), poor judgment (11.4%), inadequate lookout (11.1%), failure in
communication (9.1%), third party deficiency (8.4%), technical failure (7.9%), inappropriate and
ineffective maintenance (5.9%).

Figure 10: Immediate cases (all accident types)

18.0 17.0
16.0
14.0
12.0 11.411.1

10.0 9.1
8.4 7.9
8.0
5.9 5.8
6.0 4.9 4.6 4.5
4.0 3.3 3.0
2.7
2.0 1.9 1.6
2.0 1.2 0.9
0.4 0.4 0.3 0.3
0.0

Contributory causes of all types of accidents

In terms of the contributory causes identified by accident investigators across all accident types, the most
common were found to be: inadequate risk management (27.1%); third party deficiency (16.6%);
weather/other environmental factors (15.4%); and ineffective use of technology (15%). Failure in
communication (14.4%), inadequate training/experience (11.4%), rule violation (8.9%), and poor
emergency response (8.2%) were also noteworthy contributory causes.

12
Figure 11: Contributory causes (all accident types)

Contributory causes (all accident types)

30.0 27.1

25.0

20.0
16.6
15.4 15.0
14.4
15.0
11.4
8.9 8.2
10.0 7.8 7.5
6.2 6.2 5.9 5.3
5.1 4.2
5.0 3.9 3.8 3.6 3.2
2.9
0.9 0.6
0.0

Immediate and contributory causes of all types of accidents combined

To look at the overall picture, the immediate and contributory causes for all types of accidents were
combined and then ranked in descending order. The following were identified by accident investigators
most frequently (see Table 3): inadequate risk management; third party deficiency; failure in
communication; weather/other environmental factors; poor judgement; inadequate lookout; ineffective
use of technology; inadequate training/experience; rule violation; inappropriate/ineffective
maintenance; technical failure; poor design; fatigue, pilot error/mishandling; poor emergency response;
communication problem between master and pilot; unsafe speed, distraction; under-manning;
inappropriate use of technology; lack of manufacturer guidance; alcohol; overloading.

13
Table 3: Overall causes of all types of accident

No. of Cases %age of cases

Inadequate risk management 306 44.2
Third party deficiency 173 25.0
Failure in communication 163 23.5
Weather/other environmental factors 139 20.1
Poor judgment 133 19.2
Inadequate lookout 120 17.3
Ineffective use of technology 115 16.6
Inadequate training/experience 113 16.3
Rule violation 85 12.3
Inappropriate/ineffective maintenance 84 12.1
Technical failure 80 11.5
Poor design 72 10.4
Fatigue 71 10.2
Pilot error/mishandling 62 8.9
Poor emergency response 59 8.5
Communication problem between master/pilot 43 6.2
Unsafe speed 41 5.9
Distraction 40 5.8
Under-manning 37 5.3
Inappropriate use of technology 35 5.1
Lack of manufacturer guidance 29 4.2
Alcohol 19 2.7
Overloading 10 1.4

Conclusion

‘Inadequate risk management’ was most commonly identified as both an immediate and a contributory
cause of accidents when all types were aggregated. In some respects this is unhelpful as in relation to
very many accidents it is inevitably the case that inadequate risk management can be identified as part
of the overall picture and this does not shed a great deal of light on how accidents might be avoided in
the future. However, inadequate risk management was identified as the immediate cause of 17% of all
accidents and this may indicate that there is an underlying problem of poor training or recruitment
practice.

Overall, ‘third party deficiency’ is the second most common accident cause identified by investigators. It
is the second most common immediate cause of fire/explosion and of ‘other’ accidents. This indicates
that there is a pressing need to look beyond the operational staff within a company (on board and ashore)
when considering appropriate strategies for accident mitigation. It is arguably the case that the focus by

14
regulators and enforcement agents on third party deficiencies has been inadequate, to date, and that this
needs urgent rectification.

‘Failure in communication’ is the third most commonly identified cause of accidents. It is the most
frequently identified immediate cause of groundings and the second most common immediate cause of
collision, contact and close quarters incidents. It is often not possible to discern the underlying causes of
communication failure. In some incidents information is simply not passed on and in others there may be
language or hierarchical barriers at play. This is an area where better training and recruitment practices
would be expected to have a positive impact providing the industry with a constructive way forward with
regard to accident mitigation.

There are further areas where improved recruitment strategies and training practices could prove
effective in accident prevention. For example, in relation to incidents where poor judgement is identified
as a factor, better training or more careful selection procedures could serve to reduce accidents in the
future. Similarly, in this report, there are various factors identified as causing accidents which could be
addressed via better training including: ineffective use of technology; inadequate training/experience;
rule violation; inappropriate/ineffective maintenance; poor emergency response; unsafe speed;
inappropriate use of technology and overloading.

Finally the analysis also clearly indicates areas where management practices need to be addressed and
where it can be inferred that shore-based managers need to place greater priority on safety and less
emphasis on commercial considerations (e.g. with regard to weather routing, crewing, loading, and
ensuring that equipment manuals are readily available in appropriate languages on board).

References

Allianz (2012), Safety and Shipping 1912-2012: From Titanic to Costa Concordia, An insurer’s perspective
from Allianz Global Corporate & Speciality, www.agcs.allianz.com.
Borch, D.F., Hansen, H.L., Burr, H. and Jepsen, J.R. (2012) ‘Surveillance of deaths on-board Danish
merchant ships, 1986-2009’. Occupational and Environmental Medicine, 63(1): 7-16.
Hansen, H.L. (1996). ‘Surveillance of deaths on-board Danish merchant ships, 1986-93: implications for
prevention’. Occupational and Environmental Medicine, 53: 269-275.
Roberts, S.E. and Marlow, P.B. (2005). ‘Traumatic work related mortality among seafarers employed in
British merchant shipping, 1976-2002’. Occupational and Environmental Medicine, 62: 172-180.
Tang, L., Acejo, I., Ellis, N., Turgo, N. and Sampson, H. (2013) 'Behind the Headlines? An Analysis of Accident
Investigation Reports', SIRC Symposium, Cardiff University 3-4 July, ISBN 1-900174-46-4.

15