The photograph is of a couple of what today are quite small cruise ships, believe it or not, alongside in Barcelona. They illustrate an article I wrote some time ago on the subject of safe return to port and which I have revived due to a recent event close to the Norwegian coast involving the Viking Sky, by today’s standards also quite a small cruise ship. It has raised questions about reliability of such vessels and their suitability as a safe haven after a major misfortune on board. 

The major misfortune in this case was the failure of all four of the ship’s main engines in adverse weather when the ship was quite close to the rocky Norwegian shore. Initially it seems that the engines could not be restarted and as the ship drifted towards the rocks, beam on to enormous seas, it became necessary to implement evacuation procedures using helicopters with winching capability. This was a lengthy process and before it was completed the ship’s anchors held and its progress towards the beach was halted. Then they got the engines going and it was escorted (I think) into port. The Norwegian investigators determined that the engines had stopped due to low oil pressure even though the oil levels were “within acceptable limits” and there had been no alarms.   

 So the initial results of the investigation created more questions than answers, and many seafarers who have sailed on what were considered to be large ships years ago will continue to question the capabilities of today’s mammoth cruise ships to provide a means of evacuation, and now it seems, to provide a place of safety. So starting with the process of evacuation, how would all those passengers and crew, now sometimes exceeding 7000 people manage to get off. The Harmony of the Seas for instance carries 6780 passengers and 2100 crew, a total of 8,890 people. The ship appears to be provided with 18 lifeboats each capable of carrying 370 people. Apart from the problems relating to actually getting 370 people into a single lifeboat, just multiplying up, the maximum number to be evacuated by lifeboat would be 6660 people leaving a couple of thousand who would probably have to be evacuated by liferaft.

 But we are talking about extreme events apparently. The IMO requires that all passenger vessels more than 120 metres in length constructed after 2010 should be provided with the means of a safe return to port. This is based on the premise that the vessel itself is the safest means of survival after an major accident. We can only imagine the chaos which would result if an instruction to evacuate was given on something the size of the Harmony of the Seas from, so to speak, a standing start. I have spent about 20 years, amongst other things, analysing accidents, particularly on offshore installations of one sort or another. And some of the major findings are that many people do not know which boats to get into and even which they were in when rescued. Another major failing is that those in charge seldom give an instruction to lower the lifeboats, even if people are in them. Often it is left to the lifeboat coxswains to decide when to leave. This was particularly true of the Costa Concordia where the coxswains did sterling work. Another failing in the procedures developed for emergencies is that not enough space is provided for mustering since it is never expected that everybody who needs to muster will actually be there. Is there room, for instance, adjacent to each of the boats on the Harmony of the Seasto muster 370 people. The company would probably say no, but there is space elsewhere, and that could be true but it would be good to see how the processes have been developed and theoretically how it would all play out. In fact it is said that the owners of these large cruise ships determine that the evacuation process would work using software. Obviously the software did not consider a breakdown close to a hostile coastline, so what else did not not consider? Or is it just a whitewash anyway?

But we should bear in mind that the intention is that evacuation is an extreme event, and that actually the ships are, or will be, structured so that they can safely return to port. The accident scenarios considered are single compartment fire, and single compartment flooding with, it seems to me emphasis being placed on the fire scenarios. There is a relationship between the safe vertical zones and the accidents, and it is required that after a fire casualty there be sufficient space available within the remaining safe zones for all passengers to be accommodated. In addition it is necessary that a certain number of systems should remain available subsequent to a fire in a single space. These systems are: propulsion, steering, navigation, fuel systems, internal comms, external comms, fire main, fixed fire extinguishing, fire and smoke detection, bilge and ballast systems. In order to support the passengers, who would have to be moved from the zone containing the fire the following services should be available, sanitation, water, food, space for medical care, shelter from the environment, means of preventing heat stress and hypothermia, lighting and ventilation. And here we might remember the American cruise ship Triumph of the Seas on which there was a complete power failure, the services lost included the lavatories. One can hardly imagine the distress resulting, particularly since the thing drifted about for days until taken in tow. It was known in the American media as the “poop cruise”.

Most of the classification societies have produced guidance to help designers to develop vessels capable of conforming with the regulations and once approved they will be noted as “Passenger ship – SRTP”. But it is one thing to produce guidance and another thing to follow it, and some of the guidance provides possible distribution for propulsion systems which are as one would expect, pretty complex since it is necessary to distribute them through more than one vertical safety zone.

There are rules, or as some have defined them “performance requirements’ concerning the SRTP process, this is because the regulations require that systems remain operational but do not define what this means. And also to be taken into consideration is the distance that the ship is likely to be from port, and the speed at which it is supposed to get home. There is guidance which suggests that a speed of seven knots or half the design speed, whichever is less will be required. In addition, on the loss of any compartment within the casualty threshold, there must be sufficient fuel available for the vessel to reach port, and if there is no limit on the vessel’s range of operations  a range of 2000 miles is recommended. 

If the vessel in question is a ferry then obviously the facilities required are less than those which would be required for a cruise ship many miles from the nearest land. It is also said that some manual activities are allowed but that any manual activities to restore systems must be carried out within one hour of the accident.

Some classification societies suggest that a process similar to an FMEA (a failure mode and effect anaysis) is required for each of the disabled compartments which can amount to hundreds of FMEAs, one for each of the differing compartments for each of the required systems. Such processes are not simple. In my life dealing with oil rig safety I have had occasion to carry out FMEAs on a variety of systems. Ballast systems were a favourite for some legislations and on one rig which had a manually operated valve in the system in the pump room, at the bottom of a shaft which was only accessed by an elevator, I had a job to convince the owners that they had to include the reliability of the elevator (lift to you Brits) in the FMEA. 

We can see that some of the arrangements are fairly straightforward. In relation to the passenger space, it is a matter of ensuring that there is sufficient room for the passengers from one safety zone in all the others, and that food would be available. This would seem to make the “Burma Road” redundant; it used to be an alleyway extending for much of the length of a passenger ship just above the tank top, along which most of the provisions and meals before and after preparation were transported. 

But such things as electrical supplies are more problematical and when I used to do safety case development for oil rigs we used to visit every compartment and consider the very things the new regulations deal with, fire and flood, and for us, in addition, explosions. We would look at all the systems in the space and passing though the space, and consider the level of risk, although the criteria involved were slightly different. Typically if a space contained a load of essential electrical systems we might recommend the inclusion of a fixed fire fighting system, which would be damage limitation. But the same space on a passenger ship would require duplication or alteration since the criteria would assume that the space was burnt out, but that the systems lost in that space would still be available elsewhere.

And in all of this, as well as the whole process requiring to be developed when the ship is still on the drawing board, it is going to be down to Class, by the look of it, to decide whether the results are acceptable, since most registries nominate classification societies to act on their behalf in relation to safety. So two things. The first, is this just a blank cheque for the classification societies? And the second don’t take any of the above as gospel, it is all very complex.