Marine Incident Investigation

Module 06
 Module 06
Learning objectives
1. Analysis stage 1: Data structuring
2. Analysis stage 2: Cause analysis
 The Investigation Process

Initial Planning and Collecting Reporting/

Root-cause
reporting/ preparing Collecting and recommendations Follow-up
and cause
securing the for an evidence structuring to avoid and review
analysis
scene investigation data recurrence
 Various methodologies
Examples for data structuring Examples for Data analysis
our course our course
• HFACS - Human Factors Analysis and • Reason‘s Model (ATSB)
Classification System
• Modified Reason‘s Model (ATSB)
• MTO - HuMan – Technology - Organisation
• 5 WHY
• AcciMaps
• MAIIF / IMO Model
• Fault trees
• NI Model (modified MAIIF / IMO)
• Mind maps
• Loss Causation Model
• STEP - Sequential Timed Events Plotting
• Barrier Analysis
• Barrier Analysis
• Big Data Analysis (Correlation Analysis)

Choose the most suitable for you !

 Reminder
Incident Model
Degradation /
Routine Service Built up Period Critical Period Recovery

Event Event Event Event Event Event Event Event Incident Event Event
1 2 3 4 5 6 7 8 9 10

Latent Factors Preconditions for Active Failures

Unsafe Acts
 Methodology: Structured visualisation

Detectives office Maritime investigation office

STEP – Diagram
(Sequential Time Events Plotting)
 STEP diagram - guidelines
Events Are actions, omissions or other happenings, e.g. ‘crane strikes building’.
Should be stated simply using one noun and one active verb.
Should be quantified as much as possible and where applicable, e.g. ‘the worker
fell 6.3 metres’ rather than ‘the worker fell off the platform’.
Should indicate the date and time of the event, when they are known.

Contributing factors Describe conditions or circumstances that existed and possibly influenced or
affected the event. These factors include risk controls that failed or were missing
some of which could be safety issues, e.g. watch keeper fatigued; insufficient
manning; safe manning policy.
Must be associated with the corresponding event.

Primary event sequence Encompasses the main events of the accident and those that form the main line of
the chart.
Secondary event sequence In collisions or other complex cases a secondary event line helps depict clearly the
events and how they related in time to each other.

Source: MAIIF-Manual 2014 / [Link]

STEP diagram - guidelines

Source: MAIIF-Manual 2014 / [Link]

 STEP Diagram - Charting
Routine Built-up Critical Incident Degradation /
Period Period Period Recovery
Vertical:
Actors Horizontal: Timeline

Condition Condition Condition

Samples:
Vessel Event Event Event Event
Charting the sequence of events
Other vessel Event Event Event How did the incident occur?
Event
Master Event
Event Event
Proofing: going backwards
Pilot

OOW Event

Helmsman Event
 STEP Diagram - Sample Critical Degradation /
Routine Period Built-up Period
Period Incident Recovery
Actors 05:20 06:20 06:30 06:35 06:40 06:45 06:50 07:08 07:10 07:12
Time
Visibility
Sunrise
12 miles

Departing Pilot Course SSE Collission Speed reduced

Cargo vessel from port disembarked Speed 12 kn Ship turned

Observed AIS
Scanned Felt that something Called fishing
signals on Started
Took over horizon hit the bow vessel on VHF
OOW from Master through bridge
the radar administrative
Did not see Saw fishing vessel Asked if assistance
works
windows close along stb side is required
any targets
Command
Master of vessel
Left bridge

Trawling for prawns 6 miles from Stop trawling, Considerable

Small fishing Moving
AIS on passive mode to conceal vessels position cargo vessel starting Turned 180 damage,
astern
vessel Navigation lights on / day shapes for trawling displayed in heading hauling the
1 kn
degrees Still buoyant
line trawl net Propulsion working

Current
Skipper
Realized imminent
Noticed cargo vessel
danger Checking damage
Reactivated AIS
Set engine full astern
 Group activity 04

Create a STEP Diagram of your incident case

 Barrier analysis
The ‘Swiss Cheese Model’

Some holes are due

to active failures

Other holes are due

to latent conditions
 Barrier analysis
Bow-tie analysis
Preventive Barriers Mitigating (reactive) Barriers
C
O
H N
A

(
S L
Z E O
A EVENT Q S
U S
R E E
D N S
S C

)
E
S
Technology Process People
13
 Barrier analysis
Bow-tie Analysis
Barrier matrix - example:

Hazard Preventive How did the Event Mitigating How did the Consequence
Barrier* barrier perform ? Barrier* barrier perform ? (Loss)

*Mark the barrier:

T – Technology
P – Process
H – Human (Persons)
 The Investigation Process

Initial Planning and Collecting Reporting/

Root-cause
reporting/ preparing Collecting and recommendations Follow-up
and cause
securing the for an evidence structuring to avoid and review
analysis
scene investigation data recurrence
 The 5 WHY Tool
• The 5 Whys method is part of the Toyota
Production System. Developed by Sakichi
Toyoda, a Japanese inventor and industrialist,
the technique became an integral part of Lean
philosophy.
• “The basis of Toyota’s scientific approach is to
ask why five times whenever we find a
problem…By repeating why five times, the
nature of the problem as well as its solution
becomes clear.“ Taiichi Ohno
 5 WHY Tool
Applying the 5 WHY Tool:
1. Form a team of experts from various Problem
departments (Technical, QHSE, Marine, WHY
Operations, etc.)
Reason 1 Reason 2 Reason 3
2. Define the issue – scope/focus
WHY
3. Ask WHY Sub-Reason
Sub-Reason Sub-Reason
 Advice 1. Don’t ask too many Whys. WHY
Focus on finding the root cause.
Sub-Reason Sub-Reason Sub-Reason
 Advice 2. Sometimes there could be
more than one root cause. In these WHY
cases, the 5 Whys analysis will look Sub-Reason Sub-Reason
more like a matrix. WHY
Sub-Reason
 5 WHY Tool
Example
Sailor injured in fall on deck
WHY
He slipped
WHY
Grease on deck Running fast
WHY
Deck not cleaned after maintenance Urgent call by officer
WHY
Maintenance activities interrupted Security alarm
WHY
Urgent security issue - Stowaway
detected
18
MAIIF / IMO Model

Source: MAIIF-Manual 2014 / 4.2.5

 Modified MAIIF / IMO model
CASUALITY EVENT
(Accident)
WHY
ACCIDENT EVENT(s)
(Immediate Event(s) leading to the Accident)
WHY
HUMAN ERROR TECHNICAL FAILURE
CAUSES CAUSES
WHY
OPERATIONAL CAUSES
WHY
SYSTEM CAUSES
© Peter Werner, adapted from
MAIIF / IMO model
 Modified MAIIF / IMO Model
Level of Analysis Strategic Question

Casualty Events What major events best describe this accident/

Accident Events What (immediate) events in the sequence of events that led to the
casualty were BOTH significant and inappropriate in the context of
what was happening at the time?
Human Error and/or Technical What best describes the human error (e.g. slip, lapse, rule /
failures knowledge based mistake , wilfull violation) or other technical
failure?
Operational Causes What shortfalls in the shipboard operation allowed the errors and
failures to occur?
Management System Causes What management system related factors influenced the relevant
shipboard operations?
Adapted from MAIIF / IMO mode MAIIF-Manual 2014 /4.2.5
 Analysis process – MAIIF / IMO strategy
1. Develop a sequence of events diagram (STEP).
2. Identify which events in the sequence should be considered as Casualty and
Accident Events.
3. Analyse each Accident Event to determine why the event occurred
(discovering the background to individual actions and the Human Error and
Technical Failure).
4. For each Individual action and technical failure analyse the Operational
Causes which may have had an influence including any Risk Controls /
Barriers that failed or were not in place.
5. Analyse further to discover why the relevant Operational Cause existed to
discover the Management System Causes that underlay the events.

Adapted from MAIIF / IMO mode MAIIF-Manual 2014 /4.2.5

 Stop Rule
How far back into the system does an investigation need to go –
that is, what is the ‘stop’ rule?
If the investigation is:
• Still identifying issues that are significant and could practicably be addressed by
the company to improve the Management System, then the investigation should
continue.
• Identifying causes that the company could not reasonably be expected to address
then the investigation should be finalised.

Source: MAIIF-Manual 2014 / 4.2.5

Modified MAIIF / IMO Model - categories
CASUALITY EVENT
(Accident)
WHY
ACCIDENT EVENT(s)
(Immediate Event(s) leadintog the Accident)
WHY
HUMAN ERROR TECHNICAL FAILURE
Human Errors Causes I Technical Failure Causes
CAUSES CAUSES
WHY
OPERATIONAL CAUSES
Operational Causes

Elements of the Safety management System (Policies, Procedures, Instructions)

WHY
SYSTEM CAUSES
Compliance with Rules & Regulations / Legislation © Peter Werner, adapted from
MAIIF / IMO model
 Modified MAIIF / IMO Model - Sample
Casuality Event: Capsizing
Accident Event Human Error
Human Error Causes: Willful violation by shoreside planning

Operational causes: Lack of counterchecking by vessel

Management System: Key Shipboard operations – ISM 7

Lack of regulations – leading to new IMO
Photograph credit: L. Vanrell
resolution on control of container weights
Loss causation model – an analogy
Flu:
1. Symptoms:
• Snuff
• Cough
• Headache, etc.
2. Causes:
• Bacteria
• Viruses
• Catching a cold
3. Why does not every person fall ill ?
• Strong Immune System
• Preventive measures
 Loss causation model
LOSS
How it happened ?

Why it happened
INCIDENT Flu

IMMEDIATE CAUSES Symptoms

Substandard Acts / Substandard Conditions

BASIC CAUSES Causes

Personal Factors / Job + System Factors

Immune
LACK OF CONTROL / MANAGEMENT SYSTEM
System
 Loss Causation Model - Sample
Loss: Damage to vessel / cargo / environment
Incident Capsizing
Immediate causes: a) Substandard acts:
Incorrect stowing of containers
b) Substandard conditions:
Cargo weight unknown not verified
Basic causes: a) Personal factors:
----------------
Photograph credit: L. Vanrell b) Job / System factors
Lack of supervision
Management System: Key Shipboard operations – ISM 7
Lack of regulations – leading to new IMO
resolution on control of container weights
M-SCAT Tool of DNV
 M-SCAT - Advantages
• Structured Incident Investigation by pre-defined categories / parameters
• Easy to understand / self-explanatory
• Structured Reporting
• Data Collection
• Includes expert knowledge
• Can be modified for own operations
• Possible use for analysis of Non-Conformities
• Possible integration in Fleet Management Software Solutions
 MaRCAT Tool of ABS
Marine Root Cause Analysis Technique
 MaRCAT Tool of ABS
Marine Root Cause Analysis Technique – Details (extract)
 MaRCAT Tool of ABS
Marine Root Cause Analysis Technique – Details (extract)
 Barrier analysis
Bow-Tie Analysis
Barrier Matrix - Sample:

Hazard Preventive How did the Why did the Event Mitigating How did the Why did the Consequence
Barrier barrier perform ? barrier fail ? Barrier barrier perform ? barrier fail ? (Loss)
 Group activity 05

Analyse your incident case

Option 1: Barrier / Bow Tie Analysis
Option 2: Modified MAIIF/IMO method
Option 3: Loss Causation Model
End of Module 06
Any questions?