Kristiansen Svein. Maritime Transportation: Safety Management and Risk Analysis
Подождите немного. Документ загружается.
14.4.3 Decision Support
Regulation 29 in Chapter III of SOLAS states that a decision support system for
emergency management shall be provided on the navigation bridge of passenger vessels.
The system shall, as a minimum, consist of printed emergency plans. All foreseeable
emergency situations shall be identified in the emergency plans. In addition to the printed
emergency plans one may also accept the use of a computer-base d decision support system
(DSS) on the navigation bridge. A DSS provides all the information contained in the
emergency plans, procedures and checklists. The DSS should also be able to present a list
of recommended actions to be carried out in foreseeable emergencies. The main objectives
of a DSS include:
. Issue warnings of dangerous situations and damage
. Detect critical trends
. Give a quick presentation of critical information
. Presen tation of contingencies
. Enhance the overall understanding of the emergency situation
The emergency DSS shall have a uniform structure and be easy to use, and the
following data from sensors and alarm systems might be presented in time series:
. Draught, heel, trim, freeboard
. Water level in tanks and compartments
. Status of watertight doors and fire doors
. Temperature and smoke concentration
. Status of all emergency systems
An emergency DSS may also integrate input from operators, external sources and
static information such as hydrostatic calculations (curve sheet and stability). Figure 14.6
outlines the data structure for an integrated fire-fighting system (IFFS), which may be one
Figure 14.5. Land support team.
436 CHAPTER14 EMERGEN CY PREPAR ED N ESS
module within a more comprehensive decision support system (Kristiansen, 1994). The
IFFS may, for example, support the crew, provide information to passengers and external
parties, and be used to control remotely operated firefighting systems.
A very important requirement for all dec ision support systems is fast and user-friendly
input and output of information. The use of graphical interfaces in the DSS is therefore
highly recommended. Two examples of how such graphical interfaces might be configured
are given in Figures 14.7 and 14.8. In addition to present ing the instantaneous fire
situation, the integrated firefighting system shown in the figures may also be used for
Figure 14.7. IFFS ^ localization of fire on deck plan. (Source: Kristiansen,1994.)
Figure 14.6. The data structure for an integrated fire-fighting system. (Source: Kristiansen,1994.)
14. 4 EMERG ENCY PREP AREDNESS ACTIVITIES AN D FUNCTIONS 437
preview (prognosis) and maintenance of emergency information systems, as well as logging
of events and developments.
14. 5 HUMAN BEHAVIOUR IN CATASTROPHES
As a result of investigation of and research on catastrophic accidents we have improved
our understanding of how humans react in emergency situations. This enables us to
make realistic assessments of what can be expected in terms of evacuation effectiveness,
which again gives us a better basis for design of life-saving appliances. In particular,
investigations and research have confirmed that evacuation training is of great importance
in terms of reducing the potential consequences of seri ous accidents.
1 4.5.1 General Characterization
Maritime accidents are normally of a very dramatic nature. Some important chara-
cteristics of emergency situations are presented in Table 14.2. The emergency situation is a
function of the physical nature of the accident, the dramatic and uncontrolled
development (or escala tion) of events, and the perceived threat to people’s own safety.
The degree of drama in maritime accidents is further compounded by the degree of
isolation that is experienced on a ship and the limited availability of assistance from
external resources (e.g. other ships). The degree of rescue and salvage help is often limited
or delayed, and there is nowhere to flee other than evacuating into the sea.
The degree of drama that can be experienced onboard a ship may be illustrated by the
loss of the Ro-Ro passenger ferry Estonia (JAIC, 1997). The vessel sailed at a speed of
Figure 14.8. IFFS ^ localization of fire in section. (Source: Kristiansen,1994.)
438 CHAPTER14 EMERGEN CY PREPAR ED N ESS
about 14 knots in bow seas, with a significant wave height of 4.3 metres, when the bow
visor’s three lock ing devices failed. When the bow door fell off the locks on the inner ramp
failed, allowing water to flow into the vehicle deck. It has been estimated that the water
inflow might have been in the order of 300–600 tonnes per minute. Because of free-surface
effect the ship heeled to 30
within minutes, and simulations indicate that the vessel
reached a heel of 60
in only 16 minutes. At 40
heel the water reached the windows
on deck 4, probably resulting in progressive flooding. Owing to the rapid development of
this event, no alarm was given nor was any evacuation organized from the bridge. Later
studies indicated some local attempts to assist passengers, and that some passengers
managed to reach the boat deck, although without being able to launch any lifeboats.
Many of these saved their lives when they managed to get into the liferafts that were
released as the ship sank. Almo st immediately after the vessel started to heel, people had
problems with leaving their cabins and movement in the narrow corridors (1.2 m wide) was
Ta b l e 1 4 . 2 . Characteristics of eme rg en cy situati ons
Parameter Characteristics
Accident nature Degree of immediate threat to own life
How fast the events develop or situations change
Lack of warning People are unprepared for the next event
High degree of uncertainty
Influenced by rumours and ‘hearsay’
Time pressure Quickly changing situations
Drama Degree of injury and number of fatalities
Despair, fear and other stress reactions
Physical chaos Trapped in enclosed areas, moving objects
Blocking of escape routes
Darkness, smoke
Vessel state Fire, explosion, water ingress, sea motion
Heeling, sinking
Threat to own life Heat, lack of oxygen, drowning
Impact from explosion, structural failure
Extreme weather
Lack of control Strong feeling of anxiety
Impaired by own stress reactions and trauma
Lack of information
Lack of leadership, team spirit and solidarity
Influenced by reactions of other people
Isolation The vessel is an ‘island’ in the ocean
No or limited assistance from other vessels
or land-based resources
Critical delay in rescue and salvage
No safe haven: forced to evacuate into the sea
1 4.5 HUMAN BEHA VIOUR IN CA TASTROPHES 439
difficult. Many passengers were trapped inside their cabins, and many of those that were
able to get out of their cabins got stuck in staircases. Loose furniture and large objects
also hampered movement in public spaces. It has been confirmed by many of those that
survived that the pe ople onboard were struck by well-known emergency react ions, ranging
from panic to apathy, despite early attempts to take responsibility and assi st each other.
Of the 989 people onboard it is judged that only 300 reached the outer decks, and only 160
of these succeeded in boarding a liferaft or lifeboat when the vessel sank. In the end,
helicopters or vessels picked up 138 people, giving a survival rate of only 14%.
Studies of stress reactions under emergency situations show that it is feasible to make
a distinction between four phases of an emergency or catastrophic event/situation (Sund,
1985). These phases and corresponding stress reactions are presented in Table 14.3. The
early phases of ‘pre-accident’ and ‘warning’ are characterized by denial and/or a feeling of
being invulnerable. This may lead to a critical delay in the necessary response or fighting
of the accident. In the ‘acute’ phase people are typically subject to more dramat ic reactions
such as shock, panic or becoming paralysed. If these reactions strike the majority of the
crew and passengers onboard a ship, the consequences may be severely worsened. It is
important to note that there is a risk of developing so-called post-traumati c reactions. This
knowledge has led to a greater focus on treatment and counselling in the aftermath of
accidents and catastrophes.
Research has shown that persons involv ed in emergencies have a limited ability to deal
with challenges related to evacuation and salvage operations. As indicated in Table 14.4,
people in emergencies have a tendency to become narrow-minded and stereotypic, and
become unable to deal adequately with complicated problems. An immediate lesson of this
fact should be to design simple evacuation systems and other life-saving appliances. For
example, in a number of emergency situations there have been accidents related to the
release of lifeboats, such as inability to activate the system and premature release leading
to uncontrolled fall.
Sund (1985) has also given indicative numbers on the relative distribution of how
people manage emergencies. These are presented in Table 14.5. The group that behaves
Ta b l e 1 4 . 3 . Stress reactions in different phases of an emergency event/situation
Phase Stress reactions
Pre-accident Denial: ‘This will not happen to me’
Warning Denial, illusion of being invulnerable
Acute Shock and stress reactions: alarm, psychosomatic, passiveness,
uncontrolled behaviour
Intermediate Development of syndromes: emotionally unstable, depression, guilt,
isolation, over-reaction
Post-accident Post-traumatic disturbance such as stress and neurosis, etc.
Regaining emotional stability, control and good health
Continued need of treatment
Source: based on Sund (1985).
440 CHAPTER14 EME RG EN CY PREPARE D N ESS
optimally and takes leadership may be from 10% to 30% of the total. A larger group, of
about 50–75% of the total, will be slightly reduced but will function reasonably well with
adequate leadership. Strong psychic reactions can be seen for as much 25% of a group,
while between 1% and 3% will lose mental control and/or experience nervous breakdown.
In ad dition to the characteristics of the emergency situation under consideration, the
following background factors may determine the degree of adequate/balanced behaviour:
. Earlier experience with emergencies
. Personality type and psychic health
. Duration of employment and age
. Leadership experience
. Intelligence
Ta b l e 1 4 . 4 . Negat i ve stress react ions
Behaviour Characteristics
Sensing Narrow-minded, selective focus
‘Everything or nothing behaviour’
Cognition Stereotypic, ‘frozen attitude’
Short-term oriented, loss of perspective/overview
Unable to solve complicated problems
Reaction Limited search for information
Stereotypic behaviour
Perseverance/persistence
Impulsive or lamed
Ta b l e 1 4 . 5 . Stress react ions in eme rg en cies: d ist ri but ion with in a group
Part of population Characteristic behaviour
10–30% Behaviour balanced
Realistic perception and assessment of situation
Helps others in the group, able to co-operate
Takes leadership, demonstrates initiative
50–75% Light psychic lameness or apathy
Slightly puzzled or confused
Becomes active under leadership
No need for medical help
10–25% Strong psychic reactions needing medical treatment
1–3% Loss of mental control
Symptoms of serious nervous breakdown
Acute mental disorder or panic
Source: based on Sund (1985).
1 4.5 HUMAN BEHA VIOUR IN CA TASTROPHES 441
14.5.2 Emergency Behaviour
Considerable knowledge has been accumulated over time about concrete behaviour in
emergency situations. Reisser-Weston (1996) proposes that one should see the total
evacuation time as a function of the following phases:
. Detection
. Decision
. Non-evacuation behaviour
. Physical evacuation
The author has further proposed a task structure in emergency situations, which is
outlined in Figure 14.9. In the event of an alarm one has three basic options: act,
investigate or wait. Reisser-Weston further points out that the emergency behaviour is
influenced or determined by a set of four so-called performance-shaping factors (PSF),
which are presented in Table 14.6. Performance-shaping factors are factors assumed to
have an effect on human behaviour.
Reisser-Weston has also summarized the results of a number of studies of human
behaviour in emergency situations, primarily relating to fires in office buildings. Some of
the findings of this study are briefly presented in Table 14.7.
14 . 6 EVACUATION RISK
One of the first investigations of the risks associated with evacuation from ships was
undertaken by Pyman and Lyon (1985). The main findings of their investigation are
summarized in Table 14.8. It was found that the average probability of a fatality (or
several fatalities) occurring during an evacuation was in the order of 35%. The effect
of weather conditions was, not surprisingly, found to be considerable. Accor ding to this
study, the probability of fatalities occurring during evacuation is more than three times
Figure 14.9. Hierarchical task structure of an emergency. (Source: Reisser-Weston,1996.)
442 CHAPTER14 EME RGE NCY PREPARE DNESS
as high during hard weather conditions compared to calm or moderate weather. The
relative number of fatalities during evacuation averages 14%, i.e. one in seven trying
to evacuate does not survive. The effect of hard weather conditions on the number of
fatalities is even more dramatic: approximately five times as high in hard weather
compared to calm/moderate weather.
Ta b l e 1 4 . 6 . Performance-shaping factors (PS F) in an emergency situation
PSF Description
Structural Organization of the workplace
Physical characteristics, rules
Effective Emotional, cultural and social factors
Behaviour is affected by stress, perceived risk,
trust and cultural norms
Informational Direct information
Warning cues and information about escape routes
Communication and advice
Task and resource characteristics Possible conflict between current task or job function
and the need for evacuation
Source: Reisser-Weston (1996).
Ta b l e 1 4 . 7. Human behaviour in emergency situations: summary of research findings for fires in office
bui ldi ngs
PSF Findings
Informational High-rise office building: 14% interpreted the alarm as genuine
Informative warning system: 81% responded
Many have to observe a fire directly in order to respond
Tendency to investigate ambiguous signals further
45% were unable to differentiate fire alarm from other signals
False alarms desensitize people
Effective Investigated alarm signal: men: 15%, women: 6%
Called fire department: men: 6%, women: 11%
Got family together: men: 3%, women: 11%
Women will warn others and evacuate, whereas men have a tendency to
deal with the danger
Structural Hospitals are hierarchical: individuals respond adequately in accordance
with their position
Persons with authority are critical for mobilizing large crowds
Time to initiate evacuation without direction from staff: 8 min 15 sec
Time to initiate evacuation when directed by staff: 2 min 15 sec
In public places people are slower to break out of the normal routine
Source: Reisser-Weston (1996).
14.6 E VACUATION RISK 443
There has for quite some time been an increasing concern for the safety of bulk carriers
due to an alarmingly high number of losses and many crew fatalities . At the 70th session of
IMO’s Maritime Safety Committee (MSC), the topic of life-saving appliances (LSAs)
for bulk carriers was discussed, and it was decided to include LSAs as part of the formal
safety assessment (FSA) process for these vessels (see Chapt er 10). As a response to this,
a comprehens ive FSA project on LSAs for bulk carriers was performed in Norway (DNV,
2001), and as a part of this study the evacuation risk was estimated for this particular ship
type based on reported evacuations for the period 1991–98. The results of this study are
presented in Table 14.9.
The evacuation frequency for bulk carriers is 2.6 10
3
per ship year. The average
probability of fatality in evacuation is 29%, which is defined as the ratio of fatalities to
the number of crew at risk. This figure is much higher than the earlier cited estimate for
merchant ships in calm/moderate weather evacuation, but more comparable with the hard
Ta b l e 1 4 . 8 . Evacuati on risk wo r l d w i d e,1970^ 80
Merchant vessels
(%)
Fishing vessels
(%)
Accidents with fatality during evacuation 37 32
Fatalities among those who attempted evacuation 13 15
Hard weather accidents with fatality 78 73
Fatalities in hard weather evacuation 35 36
Calm/moderate weather accidents with fatality 16 26
Fatalities in calm/moderate weather evacuation 5 8
Fatalities among those in a fast sinking accident 86 —
Source: Pyman and Lyon (1985).
Ta b l e 1 4 . 9. Evacuat ion risk for bulk carrie rs
Type of
accidental event
No. of
events
Evacuation
frequency
(per ship year)
Fatalities No. on
board
Probability of
fatality (%)
Collision 14 3.1 10
4
116 332 35
Contact 5 1.1 10
4
54 119 45
Fire/explosion 16 3.6 10
4
6 379 2
Foundered 51 1.1 10
3
618 1209 51
Hull failure 5 1.1 10
4
0 119 0
Machinery failure 1 2.2 10
5
0240
Wrecked/stranded 23 5.1 10
4
0 545 0
Total 115 2.6 10
3
794 2727 29
Source: DNV (2001).
444 CHAPTER14 EME RG EN CY PREPARE D N ESS
weather figure (Pyman and Lyon, 1985). This supports the assessment that bulk carrier
losses often happen under dramatic conditions such as hard weather and fast sinking.
The data material from which Table 14.9 was complied has also been analysed with
respect to the effectiveness of different evacuation methods, and the results of this analysis
are shown in Table 14.10 in terms of probability of fatality. The data indicate that direct
transfer to another vessel is one of the safest evacuation means with a fatality rate of less
than 1%. It is further clear that liferafts are safer than lifeboats. It should not be a big
surprise that the least preferred method of evacuation is directly into the sea, i.e. so-called
‘wet’ evacuation.
One important piece of knowledge to be drawn from the evacuation risk data
presented above is that evacuation is a very risky activity with a high probability of
fatality. Given the high fatality rate in evacuation, it should be clear that appropriate
emergency/evacuation training and preparedness is of essential importance for seafarers.
Learned responses to critical events and situations, as well as a degree of familiarity with
simulated situations (i.e. training), can be of significant importance in terms of saving lives
in evacuation. This accounts for all phases of an evacuation, from calm and controlled
behaviour at muster stations, to correct use of personal life-saving appliances such as
Ta b l e 1 4 .1 0 . Evacuation of bulk carriers: probability of fatality for different evacuation methods
Evacuation method Number of
events
Fatalities No. to be
evacuated
Probability of
fatality
Transferred by helicopter 8 17 219 0.078
Transferred to vessel 8 1 201 0.005
Lifeboat 4 57 112 0.509
then picked up by helicopter 0 — — —
then picked up by vessel 3 24 79 0.304
unknown further salvage 1 33 33 1.000
Liferaft 3 10 68 0.147
then picked up by helicopter 0 — — —
then picked up by vessel 3 10 68 0.147
Both lifeboat and liferaft 13 81 310 0.261
then picked up by helicopter 1 21 25 0.840
then picked up by vessel 9 4 209 0.019
then by helicopter and vessel 1 5 25 0.200
unknown further salvage 2 51 51 1.000
Direct into sea (wet evacuation) 6 99 127 0.780
then picked up by helicopter 0 — — —
then picked up by vessel 4 63 90 0.700
unknown further salvage 2 36 37 0.973
Transferred to helicopter and evacuation
to survival craft
3 1 78 0.013
Transferred to helicopter and picked
up by vessel
1 0 25 0.000
Source: DNV (2001).
14.6 E VACUATION RISK 445