Sciency Words: Fault Tree

Sciency Words PHYS copy

Today’s post is part of a special series here on Planet Pailly called Sciency Words. Each week, we take a closer look at an interesting science or science-related term to help us all expand our scientific vocabularies together. Today’s term is:


The Future

The chief engineer marched herself up to the captain’s ready room, saluted, and stood at attention. The captain leaned back in his chair, reading the chief engineer’s report.

“So the starboard engine pod just exploded?” the captain said. “For unknown reasons?”

“Yes, sir,” the engineer replied. “I cannot explain it. My team has no idea what happened. The circumstances of the explosion are so unusual that we don’t even know where to start.”

“Well, couldn’t you start with a fault tree analysis?”


The Present

I first encountered the term “fault tree” in an article about the recent Space X rocket explosion (click here, it’s an interesting read). I then read more about fault tree analyses in NASA’s Fault Tree Handbook for Aerospace Applications (click here, but I’ll warn you it’s pretty dull reading).

Basically, a fault tree analysis is a method of evaluating all the things that could go wrong to produce an undesirable result (like your engine pod mysteriously blowing up). Fault trees are sort of like flow charts, and they look something like this:


As I understand it, you start from the bottom of the chart are work your way up. Which subsystems failed? How did those subsystem failures affect the main systems, potentially leading to the “top event” on your chart?

One of the key advantages to using a system like this is that it can show how two or more seemingly unimportant problems can combine to cause bigger problems farther up the tree.

According to NASA’s Fault Tree Handbook, fault tree analyses have become fairly standard for space flights ever since the space shuttle Challenger disaster in 1986. A good, well thought out fault tree can not only help figure out what caused an accident, but it can also help determine what might go wrong before an accident even occurs.

The Future

“Oh, right,” the chief engineer said, chuckling at her own foolishness. “A fault tree analysis! That’s been standard procedure since, what… the the 20th Century?”

The captain nodded.

“Sorry, Captain. Sometimes I forget I’m a real engineer and not a character in a science fiction story.”

“Happens to the best of us,” the captain said. “Carry on.”

P.S.: I have been having a really rough writing week this past week. I wonder if writer’s block can be diagrammed with a fault tree.

5 thoughts on “Sciency Words: Fault Tree

  1. The SpaceFlight101 you linked to is awesome! Not sure how I’ve managed to miss its existence.

    Interesting on the fault tree stuff. I wonder if the European space agency folks have something like that and are going through it right now.

    In IT, I know one of the problems we always have with trouble-shooting are workers jumping to conclusions about what went wrong before reviewing all the data. It often leads them to not closely scrutinize unexpected data and work on the wrong problem. I wonder if fault trees help with that.

    Liked by 1 person

    1. I really like Space Flight 101. One of my favorite things about it is that they don’t just cover NASA and ESA. They get into depth on the Russian and Chinese space programs too.

      The NASA handbook mentioned, in a casual we-don’t-like-to-talk-about-this sort of way, that NASA resisted using fault trees up until the Challenger disaster. I have no idea what the story is behind that, but I gather that a lot depended on the culture of the organization. With that in mind, I would assume ESA uses fault trees or something similar, but I can’t say for sure.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.