The intention with GBS - and in this respect especially with regard to SLA - is that the standard is an overarching and holistic approach which covers all functions and systems on board. The argument is that if there were a safety standard in place for all systems and workplaces on board, it would indirectly reflect positively on the health and safety of the crew, i.e. the OHS.
The safety level approach may be visualized by a safety knob by which IMO (indirectly through its rule-making
process) can turn/adjust the safety level – when needed – to rectify observed deficiencies. In case IMO is not satisfied with the safety - and in this specific case especially the occupational safety - they may turn the knob for increasing the safety.
However, the following questions could easily be raised:
x How safe should it be to work on board a ship?
x How many fatalities and occupational accidents do we accept before we enhance our rules and requirement?
These are indeed big questions, which are very difficult to answer. It would of course be very convenient if we had jointly agreed limits defined, but no figures are yet officially agreed upon – at least not at IMO.
The core definition of the term 'safe' means that specified, acceptable safety levels are met, regarding the risk to persons (e.g. crew, passengers), to the ship and to the environment.
However, individual safety levels for crew members and passengers on board ships have been introduced at IMO at MSC 72, 2000 [4] described in a submission concerning risk assessments and acceptance criterion.
These numbers have been used and referred to ever since and may for that reason be unofficially accepted values:
- Max. tolerable risk for crew members 10-3 annually - Max. tolerable risk for passengers 10-4 annually
The risk assessment concept or Formal Safety Assessment (FSA) was introduced at IMO some 10 years ago and has ever since been an important instrument in developing rules for the prevention of accidents on board ships. Risk assessments in the workplace followed up by guidelines (workplace instructions) for all routine functions on board have so far only been used to a limited extent on board ships, with the recently commissioned FSA study on bulk carriers being a notable exception [5].
One of the advantages in using the FSA method is that the methodology provides the possibility of determining a risk evaluation criterion. Among others, the widely used principle for determining criteria for acceptable risks is the ALARP1 principle.
The ALARP principle dictates that risks should be managed ‘As Low As Reasonably Practicable’. Both risk levels and the cost associated with mitigating the risks are considered, and all risk reduction measures should be
1 The ALARP principle, As Low As Reasonably Practicable, is defined in IMO FSA Guidelines (MSC Circ1023/MEPC Circ 392).
Human Factors in Ship Design, Safety and Operation, London, UK
implemented, as long as the cost of implementing them is within acceptable limits.
The figure below shows individual fatality risk (annual) caused be ship accidents for the crew of different ship types, including a possible individual risk acceptance criterion. The criterion consists of an intolerable risk limit (the upper line above which risk must be reduced, removed or avoided, irrespective of costs) and a negligible risk limit (the lower line below which, risks are considered to be broadly acceptable). In between is the ALARP area where risk mitigation can be considered by cost effectiveness assessment.
Figure 3: Individual fatality risk (annual) for crew of different ship types, shown together with possible individual risk acceptance criterion Source: MSC 72/16.
The boundaries shown in the figure above are based on FSA studies. What is currently lacking for formal safety assessment to be really valuable is rational risk-based acceptance criteria – safety levels.
FSA and GBS-SLA are indeed closely related to each other as they share the same objective of establishing a rational and transparent basis of safeguarding and enhancing safety and protecting the marine environment.
However, FSA is focused on being a tool following a holistic scientific method (objective, rational, etc.), whilst GBS is more focused on the structure of the regulatory system (state clear goals, state what has to be achieved, but not how to achieve it, long standing principles, independent of technology, etc.).
The GBS-SLA is to be based on controlling the risks by defining boundaries for the ALARP area and cost effectiveness criteria for safety and environmental protection. This can be achieved by either specifying an absolute level of risk (or reliability of a function/system/equipment), i.e. a single safety cut off level with any risks above that level being unacceptable and any risks below that level being acceptable, or following the ALARP principle as described in the IMO
4.1 ESTABLISHING THE GOALS
The goals (tier I) are to be based on a certain safety level.
As a start this may be done by collecting data reflecting the present level of safety in statistics.
Figure 4.: Process on how to determine the safety level Many statistics and risk assessments that focus on crew safety use the term fatality rate per ship year as common denominator. Often it is expressed by the use of the probability of potential loss of life (PPL). However, this
“unit” does not take into account occupational accidents.
It is most likely because common risk assessments are very high level analyses, which do not directly concern occupational health and safety. Another reason could be that there is no internationally adequate method on how to measure occupational accidents.
The Danish Maritime Authority has defined a strategy with regard to health and safety at sea. The overall objective for the work is that the working and living conditions – occupational health and safety – on board Danish ships has to be among the best in the world. This is achieved among other things by a continuous follow- up on occupational accidents.
Due to the intensive focus on the seafarers’ health and safety, the Danish Maritime Authority has managed to build up a significant database concerning work related accidents.
4.2 STATISTICS AND FINDINGS
A few years ago, the Danish Maritime Authority initiated a study on occupational health and safety with the purpose of finding the causes related to occupational accidents at sea.
The tables below show an extract of the study. The data covers the period 1993 to 1999 [6].
Human Factors in Ship Design, Safety and Operation, London, UK
© 2007: The Royal Institution of Naval Architects
Fatal accidents
Ship type Days at sea
Number of fatalities
Accident rate per 1,000,000 hours Container ships 3,034,183 3 0.04
Dry cargo 2,278,014 5 0.09
Coasters 3,176,915 21 0.28
Ro-ro 655,143 2 0.13
Passenger ships 2,485,879 3 0.05
Tankers 1,477,657 2 0.06
Tankers, gas 1,913,753 2 0.04
Other ships 1,511,692 5 0.14
Total 15,633,236 43 0.11
Table 1: Incidents of all identified fatal accidents by ship type.
All identified accidents Ship type Days at sea
Number of accidents
Accident rate per 1,000,000 hours Container ships 3,034,183 407 5.59
Dry cargo 2,278,014 360 6.58
Coasters 3,176,915 472 6.19
Ro-ro 655,143 181 11.51
Passenger ships 2,485,879 656 11.00
Tankers 1,477,657 335 9.45
Tankers, gas 1,913,753 78 1.70
Other ships 1,511,692 270 7.44
Total 15,633,236 2,759 7.35
Table 2: Incidents of all identified working accidents by ship type.
Classification of the ships has been done by using an official ships list (Danish Maritime Authority 1997) and a guide to all Danish ships (Dansk Illustreret Skibsliste 1997). A container ship was defined as a ship constructed for container transport and equipped with cell-guides. A coaster was defined as a dry cargo ship below 1,600 Gross Register Tons (GRT) or, if not measured in GRT, below 3,000 Gross Tons (GT). The category “Other ships”
includes a variety of different ships, such as offshore supply vessels, cattle carriers and deep sea tug boats.
A total of 2759 accidents were included in the study.
Table 1 shows the incidence of fatal injuries and table 2 shows all identified accidents. The rates in the tables are given in number of accidents per 1,000,000 hours on board. It is generally accepted that seafarers are at risk 24 hours per day (OCIMF 1997).
The empirical linkage between the fatality rate and the corresponding occupational accident are often illustrated by the “Accident-pyramid”/”Risk identification” in figure 4:
Figure 5: Accident-pyramid illustrates the perceived link between fatalities and “near misses”.
The figure describes the coherence between the numbers of fatalities and working accidents and “near miss”
situations. A logic conclusion would be that when the company culture reduces the number of “near misses”, the working and fatal accidents will follow suit. This was the philosophy behind the introduction of the mandatory ISM Code with its safety management system [8].
However, ISM and safety management system reflects only operational matters and not constructional elements like systems and functions. People working at sea should be more aware of the risks they are exposed to and through procedures avoid the incident. By nature, nobody wants to get injured. Nevertheless, seafarers continue to get injured or killed at sea.
Statistics are important to understand and prioritize the resources to where safety benefits the most. The table below contains information about which systems and functions on board contribute to the overall statistic. The Danish Maritime Authority’s accident database identifies risk factors related to ship design, which may cause occupational accidents [7]:
In table 3, all accidents in the period from 1993 to 1999 have been classified based on the activity of the victim at the time of the accident. Accidents classified under walking on deck and stairs are defined as accidents where the victim was on his way from one task to another.
On all ship types, walking from one place to another was the most frequent single cause of accidents. More than 10% of all notified accidents were caused by this activity, but they made up more than one fifth of all the accidents causing a form for disability in one way or another.
These accidents were thus in general considerably more severe than other accidents. The surfaces on decks, stairs and ladders had a major influence on many of these accidents. The placement of obstacles like tubes, handles and fittings were of importance in several accidents on deck and in the engine room. Insufficient drainage of bumpy decks caused in several cases pools of water, which especially in frosty weather were direct causes of accidents. Poor lighting was identified as another risk factor of importance. Another 20% of the serious accidents causing disability also took place on deck, stairs and ladders, but were recorded as part of another task. Also in several of these cases, the construction and maintenance played an important role.
As accidents involving passage from one location to another, ergonomics and safety concerning access-, passage and stairways may have to be evaluated further.
Gangways, particularly on smaller ships, were a cause of several serious and fatal accidents. Loose gangways caused accidents when they were rigged and taken in.
While in use, there are examples of tipping gangways and seafarers who fall off the gangway. Several of these accidents were fatal.
Human Factors in Ship Design, Safety and Operation, London, UK
Working situation at time of accident
Fatal accidents (percentage of total)
Reported accidents not causing death (percentage of total) Work on deck
Clearing up and cleaning on deck and in holds
4 (15%) 51 (2.9%) Handling of general stores 0 32 (1.8%) Lashing and unlashing of
cargo
0 67 (3.8%)
Loading and unloading cargo 3 (11%) 118 (6.7%) Mooring and anchoring
operations
1 (4%) 100 (5.7%) Preparing the ship for a
voyage
0 21 (1.2%)
Opening and closing of hatches and bow ports
0 46 (2.6%)
Rigging and taking in gangways and pilot ladders
2 (7%) 39 (2.2%) Routine tasks on deck
(controls, daily routine jobs)
1 (4%) 51 (2.9%) Maintenance on deck 2 (7%) 102 (5.8%) Painting including preparation
for painting
0 27 (1.5%)
Repair work on deck and accommodation
0 71 (4.0%)
Specialised tasks on off-shore vessels and tugs
2 (7%) 30 (1.7%)
Tank cleaning 0 34 (1.9%)
Total, work on deck 15 (46%) 789 (44.9%) Walking from one place to
another
Walking in accommodation and galley
0 33 (1.9%)
Walking on deck and in cargo holds
2 (7%) 53 (3.0%) Walking in the engine room
and repair shop
0 13 (0.7%)
Walking on stairs in the accommodation
0 31 (1.8%)
Walking on stairs and ladders on deck and in cargo holds
0 26 (1.5%)
Walking on stairs and ladders in the engine room
0 14 (0.8%)
Walking on gangway (to and from the ship)
2 (7%) 12 (0.7%) Total, walking from one place to
another 4 (14%) 182 (10.4%)
Other functions
Boat and fire drills 2 (7%) 28 (1.6%)
Duty on bridge 0 7 (0.4%)
Transport, ashore (on duty) 1 (4%) 7 (0.4%) Maritime disasters 1 (4%) 4 (0.2%) Accidents while off-duty
ashore
4 (15%) 71 (4.0%) Accidents while off-duty on
board
0 59 (3.4%)
Violence from passengers, piracy
0 17 (1.0%)
Other accidents, poorly described accidents
0 20 (1.1%)
Total, other functions 8 (30%) 213 (12.1%)
Among the most serious accidents are those related to mooring operations. Traditional mooring winches caused major accidents as the seafarers are very close to very strong tensions. This study reveals several cases where seafarers have been killed and other severely injured, including loss of legs due to bursting mooring ropes.
Handling of stores, especially in the galley, but also in other ship departments, causes accidents, as well as more permanent working disability. Poor access to stores rooms is a risk factor for accidents.
Very highly placed wheelhouses on especially smaller container ships were identified as a risk factor for accidents. Due to the height, accelerations may be very high and therefore in practice causes accidents. Lack of a sufficient number of handles played a role in some cases.
Boat drills is a well known cause of serious accidents and also in this study several serious and fatal accidents were identified. Insufficient knowledge on how to operate the systems was a major problem, but also the technical construction enabling human mistakes to be made was identified as an important factor.
Why are such statistics so important?
The answer is perhaps too obvious: we establish our own goals or acceptance criteria based on the findings.
Experience from analyzing such data indicates what is relevant to focus on.