What is a fault tree analysis?
A Fault Tree Analysis (FTA) is a systematic approach to problem-solving, troubleshooting and identifying a failure’s root cause using a diagram. A fault tree analysis can either be used to explore a single failure or systematically examine a group of components, which makes it a versatile tool for a root cause analysis.
Similar root cause analysis methods include the dependence diagram, the reliability block diagram and Markov analysis. FTA is often compared to FMEA, but there are substantial differences between the two. We’ve broken down the differences between FTA and FMA for you right here.
What is the purpose of a fault tree analysis?
- diagnose the root cause of a failure
- understand how the system can break down
- determine inherent risks
- identify measures to minimize risks
- estimate the frequency of safety breaches
What are the advantages of performing a fault tree analysis?
- improve compliance with safety regulations
- map the correlation between failures and subsystems
- establish priorities for the system as a whole
- implement changes in product or system design to minimize risks
- carry out a probabilistic risk assessment (PRA)
How to create a fault tree analysis diagram
Your starting point for a fault tree analysis diagram is the failure itself. From that top event onwards, the diagram should grow following a logical sequence – until you determine root cause.
The fault-tree analysis diagram applies boolean logic. It employs symbols to represent each event that might have caused or contributed to the breakdown, including external and conditioning events. The gate symbols (“and”, “or”) that connect them establish the relationship between input and output events. Here’s a list of all the symbols in an FTA:
You can also structure the diagram using a reliability software, in which gate symbols equal 1 or 0. Most software will integrate the information with statistical probabilities, which turns FTA into a quantitative method.
Because of these characteristics, an FTA is a common method for safety assessments and probabilistic risk assessments (PRAs). A PRA is a systematic approach to risk and reliability analysis, which estimates risk, the likelihood of failure, and the magnitude of the consequences.
When to use FTA in maintenance
We’ve already established that an FTA can be used for a PRA, which makes it appropriate for high hazard industries, including aerospace manufacturing, nuclear, chemical, petrochemical, and pharmaceutical industries. In software engineering, FTA is a cause-elimination technique for debugging.
As a curiosity, NASA preferred an FMEA analysis for the Apollo missions – which took the Man to the Moon for the first time – because the probability of returning safely to Earth was too low according to an FTA. After the Space Shuttle Challenger disaster in 1986, which disintegrated only 73 seconds after liftoff, NASA began using a combination of FMEA and FTA analysis.
Now, let’s get straight to the point…
How do you use FTA in maintenance?
When there’s an unexpected breakdown or a failure that almost leads to one, it’s good policy to perform a fault tree analysis to get to the root cause. Otherwise, a failure will happen again. For example, if the fire protection system fails, there are two possible failure modes: either (hypothesis A) a failure occurred in the fire detection system or (hypothesis B) the fire suppression system failed.
If the fire detection system failed, it means that the fire detectors failed and so did the heat detectors (both mechanisms must fail). On the other hand, if it was the fire suppression system, it means that there was not enough water in the system or that sprinklers were blocked (any of the two would be enough to cause failure).
If we conclude that the problem was not enough water, then we’ve found our root cause. Your fault tree analysis may stop here because you’ve found the “base event” (marked with a circle).
However, as a maintenance manager, you might wonder why the water pump failed. For example, whether it was lack of maintenance, wear out or insufficient capacity for the building’s needs. You can either continue with the diagram or turn to an alternative method, like the 5 Why’s analysis.
Investigating the root cause is, ultimately, what allows you to make the right changes in the maintenance plan, implement new safety rules, and assess risk. Consequently, both the asset’s availability and reliability increase over time. And this is how you can use fault tree analysis in maintenance.
Although it is not used systematically outside the industries we mentioned, the fault tree analysis is a very useful tool to determine root cause(s) and improve any company’s maintenance strategy. You can use it almost in any context, from simpler failures to more complex systems. Plus, it showcases how several events might contribute to the outcome.
What are the limitations of fault tree analysis?
There is not a single foolproof root cause analysis. FTA too has its limitations:
- It’s a unidimensional model. FTA doesn’t take into account time or the asset’s useful life, which may be a problem when you’re in the product development stage.
- FTA is a binary system. Each hypothesis is either validated or not, making it roo rigid for assets with conditional failures (failures that only happen under certain conditions, i.e. low temperatures) or partial failures.
- It is not always possible to determine the probability of failure, which invalidates FTA as a quantitative method.
Do you want to know a few alternatives for FTA? Check out our article about root cause analysis tools. The right match for you is out there!