Common Cause Failures

Question 1

Common Cause Failure (CCF) is when________, which means that the success of the system becomes uncertain.

Answer 1

Common Cause Failure (CCF) is when two or more items fail within a specified time, which means that the success of the system becomes uncertain.

Question 2

Give some examples of pre-operational root causes

Answer 2

• Design
• Manufacturing
• Construction
• Installation
• Commissioning errors

Question 3

Generally there are four types of root causes, these are:

Answer 3

Hardware,
Human,
Environmental,
External.

Question 4

To avoid CCF we use defence strategies such as:

Answer 4

Separation, 
Diversity, 
Fail-safe designs, 
Staggered testing,
Functional testing,
Proof testing, 
Quality Control.

Question 5

What are the 7 Modelling CCF Steps?

Answer 5

1. Develop a system logic model, such as a fault tree
2. Identify relevant CCF component sets
3. Identify relevant root causes and coupling mechanisms
4. Assess the effectiveness of the CCF defences
5. Establish explicit CCF models
6. Include implicit CCF models
7. Quantify the probability of system success and failure to interpret the results.

Question 6

In engineering, CCFs can lead to loss of _____, which can be challenging to identify.

Therefore, _____ models can be used as tools to help organise thinking.

Answer 6

In engineering, CCFs can lead to loss of SYSTEM FUNCTION, which can be challenging to identify.

Therefore, QUALITATIVE models can be used as tools to help organise thinking.

Question 7

CCFs are considered within risk modelling to ________.

Implicit modelling uses a CCF modelling approach that is ________ based.

Answer 7

CCFs are considered within risk modelling to estimate, prioritise, and mitigate risks properly.

Implicit modelling uses a CCF modelling approach that is probability based.

Question 8

What is Poisson Modelling?

Answer 8

It is the simplest model of component failure, and has:
• A “common shock” interpretation
• The time between shocks is assumed to be exponentially distributed
• Different shock processes are independent and display different rates.

Additionally, the rates of different processes can be added using Superposition.

Question 9

What is Beta Factor Modelling?

Answer 9

Focuses on the proportion of component failures due to common shocks.

Common cause shocks are said to force all affected components to fail, while other shocks cause “independent” component failure.

It is simple, has a small number of parameters, but is also conservative and thus doesn’t reward different levels of redundancy.

Question 10

What is Binomial Failure Modelling?

Answer 10

This model uses three parameters, namely:
• Independent shocks which occur with rate (𝜆)
• Common cause shocks which occur with rate (𝜇)
• Following common cause shocks, the components will fail independently with probability (𝑝)

Question 11

What is Alpha Modelling?

Answer 11

The failure rate depends on the group size.
It has a low number of parameters which are easy to estimate, however, physical interpretation can be less obvious.
• Parameters which give the failure probability 𝛼1, 𝛼2,…𝑛
• Group size 1,2,…𝑛
• Overall rate of failure 𝜆

Question 12

Give advantages of Beta Factor Modelling

Answer 12

• Conceptually simple
• Small number of parameters
• Conservative probability estimates
• Widely used in industry

Question 13

Give disadvantages of Beta Factor Modelling

Answer 13

• Doesn’t reward different levels of redundancy

* If independent failure rate changes, the common cause rate changes too