e5.1 - Theories/models and use of loss causation techniques

Question 1

Incident

Answer 1

An unplanned event which has the potential to cause loss of some kind

Question 2

Accident

Answer 2

An unplanned, unwanted event which causes loss of some kind

Question 3

Heinrich (1931)

Answer 3

Tip of triangle - major accident

Middle of triangle - minor accident

Base of triangle - non-injury accident

Question 4

Accident ratio studies demonstrate…

Answer 4

Most incidents do not result in a high severity outcome

Matter if chance whether an event results in injury, damage or near miss

All accidents and incidents should be investigated to find cause

Resources should be based on potential loss rather than actual loss

Question 5

Near Misses (Accident ratio studies)

Answer 5

Near misses often have an identical root cause to serious incidents and can reveal management system failures before serious incidents occur

Not every near miss or minor accident could lead to a serious one.

Question 6

Limitations of accident ratio studies

Answer 6

Different triangles from different organisations cannot be easily compared

Certain amount of data needed before comparisons

Similar injuries, different causes

Many injuries are musculoskeletal which could not have ended up as fatalities

Measures failure rather than success

Question 7

Domino Theory

Answer 7

Single cause theory

Accident occurs due to five sequential factors

Each factor is caused by the preceding one

Remove any of the first three factors, prevent the accident

Question 8

Domino theory - 5 factors

Answer 8

LACK OF CONTROL (inadequate system, standards)

INDIRECT CAUSE (personal factor & job system factor)

IMMEDIATE CAUSES (substandard act, practice or conviction)

ACCIDENT EVENT (event or contact)

INJURY/DAMAGE (the result)

Question 9

Domino theory limitations

Answer 9

Simplistic

Reactive not proactive

Not useful in prediction the likelihood of accidents

Encourages focus on immediate rather than underlying causes

Restricts the search for multiple causes

Question 10

Multi Causality theory

Answer 10

More than one cause of the accident, not just in sequence but occurring simultaneously e.g. Fuel+oxygen+source of ignition = Fire

Each must come in the right quantities at the right time

Question 11

Multi causality theory - Advantages

Answer 11

Helps to consider many possible organisational and cultural issues

Encourages search for underlying/root causes

Underlying root causes encourage wider look at org. arrangements

Question 12

Multi causality theory - limitations

Answer 12

More complex, high competence needed

Time consuming

Over complicating when causes are simple to understand

Question 13

Immediate causes

Answer 13

Unsafe act/omission (people) or unsafe condition (place, equipment, materials)

Question 14

Underlying causes

Answer 14

Management failing that allows the immediate causes to occur (e.g. failure to carry out risk assessment)

Question 15

Root cause

Answer 15

Deeper management failings from which all underlying causes stem (e.g. lack of resources)

Question 16

Swiss cheese model (James Reason)

Answer 16

Risk management in 4 layers

A failure in one layer (poor control) can be picked up by the next layer which has no holes (i.e. the control works)

Question 17

Swiss cheese model - latent failures

Answer 17

Contributory factors in the ‘system’
May have laid dormant for a long time
Introduced by organisational or managerial factors
Accumulate over time
Lack of management control
Lie dormant until triggered by ‘active’ failure

Question 18

Swiss cheese model - Active failure

Answer 18

Due to human error (accidental) or violation (deliberate)

Question 19

Fault tree analysis

Answer 19

A graphical technique used in risk assessment to provide combinations of possible occurrences that can result in an undesired result.

Used in multi causal investigation following back to its root cause using a logic diagram

Determine root causes by beginning with unwanted event (incident/accident)

Question 20

Event tree analysis

Answer 20

Based on binary logic (component either succeeds or fails)to assess protective systems designed to prevent

Analyses the possible consequences from failures

Event trees start with an INITIATING event and work towards the TOP event, the opposite to fault tree analysis.

Question 21

Event tree analysis - Limitations

Answer 21

Accuracy of data to determine probability of an event and the high degree of competence and expertise needed to complete them

Question 22

Bowtie Model

Answer 22

Method for visualizing hazards, barriers and control measures

Analysis of hazard and consequence to incidents (Known as TOP event)

Question 23

LEFT SIDE of bowtie

Answer 23

Comprises a fault tree

Barriers in place should be sufficient to stop event

Question 24

RIGHT SIDE of bowtie

Answer 24

Comprises event tree

Mitigation barriers to minimise consequences of incident