Introduction
In this short article, I would like to discuss how Fault Tree Analysis (FTA) is applied before or after an event (typically a disaster scenario), or rather Proactively or Reactively. We touched on this briefly in our blog Fault Tree Analysis Explained, but we delve into more detail and examples in this article.
Many people use FTA, and depending on where you work, and what you are trying to achieve, you will have a different take on how FTA is applied. For those new to FTA, this can be confusing, as you may be unsure which perspective you should approach FTA for your own applications. The two perspectives are:
- Reactively: using FTA to understand the causes that led to an event that has already occurred.
- Proactively: using FTA to understand the causes that could lead to some event that has not yet occurred.
It all seems pretty straight forward so far… let’s look at some examples.
An example of Reactive FTA
In 2015, SpaceX Falcon 9 experienced a rocket explosion. Elon Musk then posted the following on Twitter:
In this scenario, it is clear that Elon Musk and his team were planning on using FTA to identify the root cause of the explosion. This is an example of using FTA reactively; investigating the event after the fact. Using FTA in this way typically doesn’t involve numerical and statistical analysis.
An example of Proactive FTA
We don’t always have to wait until something has happened to identify its causes. This leads us to the proactive use of FTA.
I’ll use examples of this from my own experience, where I have used FTA to understand what the causes are of the following events, before they occur:
- Oil well fire
- Electrocution from an aircraft engine
- Failure of a safety critical PCB
- Environmental disaster caused by oil leakage
In these scenarios, I was investigating the probability of each of these events happening, before they have ever occurred (and hopefully they never will). The specific aim in each of these cases was to understand the weak links in the current system design, and identify if the level of risk is acceptable.
Typically, when performing FTA proactively, we are interested in quantifying the probability of the event in question. This is not always the case however; we could also just build a qualitative fault tree (no numerical inputs) and get lists of combinations of failures to investigate (typically referred to as cutsets).
If you aren’t familiar with cutsets, check out our blog What is a cutset in Fault Tree Analysis? for more info.
Summary
In this short article I just wanted to highlight the differences between how some people use FTA compared to other. It can be especially important to be aware of this when you are working in a larger organisation and promoting the use of such tools.
