RCA as a knowledge practice

Do we take sufficient time and trouble to ask questions to really learn from accidents, failures and near-misses? Root Cause Analysis (RCA) is one approach.

Every time things go wrong there is a huge opportunity to learn, redesign and improve performance. Bill Corcoran suggests we start with these questions:

Eight Questions for Insight:

1. What were the most important consequences? Actual, expected, potential?
2. What makes this event significant?
3. What set them up for it?
4. What triggered it?
5. What made it as bad as it was?
6. What kept it from being a lot worse?
7. What should be learned from it?
8. What should be done about it?

Some other questions:

• What is it about the way business is done that makes events like this inevitable?
• What were the earlier, better, safer, cheaper ways that we could have found the causal factors of this one without having the event?
• What's wrong with this picture, i.e., the news story?
• What do you think were the non-consequential precursors to this Real McCoy?
• Can you think of some negative causes for this one?
• Which accidents in your industry does this remind you of?
• Which related accidents does this remind you of?

To enable the learning, we need an environment where failures are celebrated (as learning opportunities), messengers are not shot, reflection in community is encouraged, there is commitment at all levels to avoid known pitfalls, share experience and learning is supported.

Adopting a reflective practice, is a useful additional regular activity within any community of practice, it helps to surface patterns and anti-patterns that capture group expertise and experience in a common advanced shared language.

I found this volume expensive and useful, (get a copy from your library)

Comments

[Ashley Bowers]

To learn from ones mistakes is always important but to celebrate them to me is not!

[cindy]

IMHO, RCA should not be just for near-missed, failure etc. It should be a standard procedure at the end of any project. It we do that it will become part of the process (of any project or happenings) therefore 'blame' would not be the focus of RCA.

Even the most perfect 'project' can be improved.

[Harold Jarche]

Sounds very much like aviation safety investigations. The aim is to learn, not cast blame.

[Shawn Callahan]

This post is timely for me Denham. I'm working on a narrative approach to lessons learning at the momemt and coming to the conclusion that the logical-scientific approach of attempting to eek out the causes of a mistake only reinforces defensive reasoning. This in turn masks there real causes.

While I think it is important to answer these questions posed above I think the group needs to revisit the stories which describe what happened. Through the detail generalities emerge and evidence (anecdotal) is available.

I've just finished reading a couple of interesting papers which explore the idea of learning from failure. Both come to the conclusion that it is very difficult to do because of defensive behaviuor. But both papers assume a logical-scietific investigation. You may have seen what I wrote about the narrative vs. logical-scientific (http://www.anecdote.com.au/archives/2005/12/balancing_narra.html).

1. Baumard, P. and W. Starbuck, H., Learning from Failures: Why It Might Not Happen. Long Range Planning, 2005. 38: p. 281-289.

1. Argyris, C., Teaching Smart People How to Learn. Harvard Business Review, 1991. May-June.