Accident Prevention Using Failure Analysis

Mitts · Feb 27, 2011

I, like most of this hiking community, have been interested with accidents that happen in the places that I enjoy to spend time playing in. I have come to believe that these accidents are very very rarely the result of a single decision or event, and are more often the result of a series of events and decisions.

These beliefs (and reading Laurence Gonzales' "Deep Survival") have led me to adopt a "red flag" approach to my decision making while hiking. This being making a mental note of situations that might domino and eventually lead to an accident. These red flags could be as simple as not getting as early of a start as planned, to noting a partner's slurred speech or stumbling step. At some point after I have noted enough red flags, I make a decision based on the sum of the things that I've noted throughout the trip, not just the latest negative event.

I recently wrote a paper about Andrew McAuley's fatal kayak trip between Australia and New Zealand, and analyzed some of the decisions he made both before, and during his trip that may have negatively affected the outcome using FMECA. I have long wondered about the feasibility of applying this approach to hiking in the Northeast as a way to develop a set of procedures that could help our community better understand the series of events that lead up to an accident. The paper I wrote, which (hopefully) explains the process more clearly than I have in this post can be read Here.

I'm posting this here in the hope that some of you might be interested in collaboratively developing this analysis for the benefit of our community. I'm also interested on any input or discussion about whether or not this approach would yield realistic information that would be helpful to the average Northeast Hiker. Any and all input is appreciated

-Andrew

Dr. Dasypodidae · Feb 27, 2011

Good stuff, Andrew.

You might also wish to contact Jed Williamson at Dartmouth (I believe that he is still affiliated there), who edits the annual Accidents in North American Mountaineering for the American Alpine Club, as he may know of similar efforts to apply FMECA to mountaineering accident scenarios.

Also, there are a bunch of scientists and engineers on this site, who I am sure will be interested in your project and be posting in this thread.

hikerbrian · Feb 28, 2011

Mitts said:
I have come to believe that these accidents are very very rarely the result of a single decision or event, and are more often the result of a series of events and decisions.

I could not possibly agree with you more!!! I have never done a formal analysis or tried to reduce this adage to a measureable metric, and I personally don't believe there would be value in doing so. (Then again, I haven't read your linked paper yet either.) IMO, as often as not, life happens outside of the metrics, and the second you decide to rely on a formula, you'll find out just how important the exceptions are. So, FWIW, I think communicating metrics to the average hiker, beyond "be aware that it's usually a series of problems that leads to the Big Problem" would do more harm than good.

All of that said, my number is 3. 3 major problems/bad decisions and you're screwed. Last weekend, for example, a group of us hiked above treeline into deteriorating whiteout conditions (1), then eventually lost the cairned trail (2). You can believe I was on high alert then. My feeling then (and now) was, one more screw up - bad ankle sprain, puncture wound from a crampon, group gets separated, take a bad route down, hypothermia - whatever, and things will get real bad real fast.

weatherman · Feb 28, 2011

That would be a very nice study. Get a panel of experts (us?!?!?) to make a list of, say, 12-15 major bad things that are controllable (since we can't do much about, say, gravity) and find 50 accidents for which reliable data are available. Get 3 raters to independently analyze each accident, then get together and see how many factors are involved in each one. Find a statistician to do regression analysis and inter-rater reliabilty, then find a journal that would publish it. Cool stuff.

Of course, you run the risk of finding a conclusion like "don't hike above treeline in the dark in the winter, duh!"

Weatherman

TCD · Feb 28, 2011

You could also continue from there using the ISO 14971 Risk Management approach. After you use the data to prepare an initial FMEA, you could identify risk mitigation strategies, such as planning and training, and then identify the residual risk (the residual risk is basically the unavoidable risk that remains even after you have conducted risk mitigation).

A presentation like that might be of interest to insurers and legislators.

Mitts · Feb 28, 2011

Some clarification

Perhaps I should clarify a few points...

My goal would not be to generate a decision tree that one would laminate and take hiking. That would be unrealistic. My goal would be to use the failure analysis to look at the act of hiking and gain insight into how different hiking "systems" are interconnected. For example: I have always wondered how much of an effect air temperature actually has on increasing the risk of a hike. An analysis might shed some light on the difference in risk between a 5 deg hike, and say a -20 deg hike.

I don't think the project would make hard and fast suggestions like "dont hike above treeline at night". It may however, allow one to compare the risk of different activities. For example: hiking Mt. Monadnock mid week alone has a similar amount of risk as hiking above treeline at night with a large, well equipped group.

But like I said earlier, this is just an idea I wanted to throw out to the community, and so far it seems like there is at least some interest in the project. Let the discussion continue!

-Andrew

Creag Nan Drochaid · Feb 28, 2011

Quantified risk

Such research might be of interest to legislators, all right. It tends to quantify what most of us know by learning from mistakes, ours and other people's.
Let us recall that the ongoing issue of lost hikers not paying for their own rescue has resulted in the legal threshold for being charged for your rescue going from you acting recklessly to you acting in a manner that is only negligent, and that within the last two years. This kind of quantitative study adds support to the experiential and intuitive mindset that counsels prudence in our outdoors activities.
A word to the wise is sufficient, most of us live to learn from our mistakes, or other people's ( I turn first to the accidents column in Appalachia), and the value of the proposed study may come when we do use it to assess residual risk in order to determine whether a rescuee was negligent. I expect that this is what the rescue review board already does after rescues, perhaps the method suggested would allow a statistical comparison among rescues after it had been used enough times and enough examples of it accumulated. Hard to say...
Creag nan drochaid

weatherman · Mar 1, 2011

Mitts said:
It may however, allow one to compare the risk of different activities. For example: hiking Mt. Monadnock mid week alone has a similar amount of risk as hiking above treeline at night with a large, well equipped group.

Intriguing.

Risk (death)=B1(1/temp in K)+ B2(wind speed) + B3 (vertical feet climbed) + B4 (1/available light in minutes till sunset) + B4 (1/coefficient of friction on ground, assuming Vibram soles) + B5(1/maximum blood flow through coronary arteries (ml/min)) + ... + X (gigantic fudge factor).
Back to work.

Mitts · Mar 1, 2011

A Call and a Raise

I will see your formula and raise you a spreadsheet

https://spreadsheets.google.com/ccc?key=0AslUe7uAR-G7dExzNWY3Y0lPR2p6d3I4MEt1eGE3TGc&hl=en&authkey=CPeRx7wC

The System requirements are things that a human needs to remain alive: cant be frozen, cant be immobile, and you cant have your head cracked open.

The External variables affect the system requirements (in the case of hiking usually negatively) and are unable to be controlled by the human: weather, terrain, etc.

The internal variables are characteristics of the system that reflect back on the system in some way: a person who has been hiking for 12 hours without food (is tired) will not generate as much heat as an "un-tired" person, and the core temp system requirement would be lower

The Equipment Mitigation Category are ways that equipment might change the affect of the external variables on the system: a down jacket would modify the air temperture's effect on core body temperture

The spreadsheet is obviously lacking in a few large ways and the feasibility of making this information quantitative is, to say the least, a little daunting. Gotta start somewhere though, right?

If anyone has any interest in adding to or modifying this spreadsheet, PM me with a gmail account user name and I will be happy to add you to the "allowed to edit" list

-Andrew

weatherman · Mar 2, 2011

Nice. What about variables that are part of several categories and can affect the system in multiple directions? The item "brain" comes to mind. Necessary for survival, mobility, and an internal factor for sure; however, it can affect the system positively through good decision-making and excellent coordination or negatively through clumsiness or dumb choices. Or all of these!!

sierra · Mar 3, 2011

Not to impune your work, but I see little value in these type of studies in relation to preventing future mishaps. There are two main factors that go into accidents in the backcountry. One, not having sufficient knowledge of mountain climbing and all it emcompasses, two, having all the above knowledge and ingnoring it in critical situations. There have been countless articles, publications, websites, all devoted to this topic and alas, people still fall victim to peril. Hey knock yourself out, but in the end, espechally considering the influx of newbies to the backcountry, you will continue to have plenty of new data to play with, imo.

Little Rickie · Mar 3, 2011

Are you an engineer?

Maddy · Mar 3, 2011

sierra said:
Not to impune your work, but I see little value in these type of studies in relation to preventing future mishaps. There are two main factors that go into accidents in the backcountry. One, not having sufficient knowledge of mountain climbing and all it emcompasses, two, having all the above knowledge and ingnoring it in critical situations. There have been countless articles, publications, websites, all devoted to this topic and alas, people still fall victim to peril. Hey knock yourself out, but in the end, espechally considering the influx of newbies to the backcountry, you will continue to have plenty of new data to play with, imo.

AGREED!
Have been thinking a lot about the lessons of life recently and my conclusion is this. Not many seem to learn from other people's mistakes, and too often we don't appear to learn a whole lot from our own blunders.

I'm sure we have all read the quote re: advice.
"Wise men don't need it, and fools won't heed it".

Perhaps it's because to get through life, most especially a life filled with great adventures, we need to go forth with the firm conviction that nothing really bad will happen to us. We convince ourselves that we are stronger, more able to handle risky situations than "so and so." We firmly believe we will have a good outcome, and if we don't we can deal with it.

Some carry on accepting the fact that their behavior is risky but the thrill, the adrenaline rush, is worth it. After all...what are the chances?

The Banff Films prompted me to reflect on this a whole lot recently.
I haven't ordered my "wing suit" yet.

I just read that 3 people from the Beserk crew were lost at sea in Antartica.
I purchased a film that was not shown, A Life Ascending, loved it, but the bottom line was, despite the fact that the guide had tons of experience, seven lost their lives in an avalanche in remote BC. It was a perfect day and there was no reason to believe anything like this would happen. The guide was very cautious checking snow for potential avalanche risk on every expedition. He was a no nonsense kind of man.

Perhaps it goes back to what Alex Lowe said.
"Remove the risk and you remove the challenge.
Remove the challenge and you wither on the vine."

Is it possible that the real choice is: "Pay for your rescue, or wither on the vine?"

When all is said and done, I think this applies across the board to all walks of life. We do learn some lessons from experience, but sometimes life has to knock us down really hard before we really pay attention, and maybe not even then.

I think a study would be very interesting and I would like to read the results, but I don't think it would be instrumental in changing most people's behavior. If you have the time and energy, it might be worth the effort to try to make a difference. After all we don't really know until we try.

Maybe SAR (F&W) could use it as a guide to promote "Hike Safe".

Mitts · Mar 3, 2011

Little Rickie said:
Are you an engineer?

Does it show?

sierra said:
Not to impune your work, but I see little value in these type of studies in relation to preventing future mishaps. There are two main factors that go into accidents in the backcountry. One, not having sufficient knowledge of mountain climbing and all it emcompasses, two, having all the above knowledge and ingnoring it in critical situations. There have been countless articles, publications, websites, all devoted to this topic and alas, people still fall victim to peril. Hey knock yourself out, but in the end, espechally considering the influx of newbies to the backcountry, you will continue to have plenty of new data to play with, imo.

I totally agree that no amount of studying or analysis (of which there has been much done) will have any sort of significant impact on how others make their decisions. My motivation comes from a desire to better understand how accidents occur so I can make better decisions for myself. Frankly, I don't care what others do.

Maddy said:
When all is said and done, I think this applies across the board to all walks of life. We do learn some lessons from experience, but sometimes life has to knock us down really hard before we really pay attention, and maybe not even then.

I think a study would be very interesting and I would like to read the results, but I don't think it would be instrumental in changing most people's behavior. If you have the time and energy, it might be worth the effort to try to make a difference. After all we don't really know until we try.

Maybe SAR (F&W) could use it as a guide to promote "Hike Safe".

Personally, I don't always learn from lessons or when "life knocks us down really hard" and I think that can be a dangerous way to learn lessons in the mountains. If I can learn how to better make decisions while hiking by thinking a situation through before it happens, then I would much rather do that than learn something "the hard way".

We all do this to some extent when we check the weather before we leave for a hike. We take in information and use it to think through how it will affect our decisions. This study would only be a more in depth way to "think through" different hiking scenarios in an effort to better understand those different scenarios.

Read above for my comments on how the study would not affect the general population. I don't see this study as a supplement to or replacement of "hike safe".

I see a lot of discussions on this forum as well as other places online and in print where hiking accidents are discussed with allot of opinion. This study, analysis, conversation, or whatever you want to call it is merely an attempt to add a little bit of data to those discussions. It is quite possible that what I am proposing simply wont work at all

Maddy · Mar 4, 2011

I was thinking more of it as a ''compliment" to the Hike Safe program.

I would very much like to read the results. Opinions are not necessarily a bad thing. We often learn something when someone gives us their "opinion", which is often derived from their own personal experience and/or education. I thing to compliment that with a formal study would be useful, at least for some folks.

I admire your enthusiasm and willingness to share the info.

Little Rickie · Mar 4, 2011

Mitts said:
Does it show?

I haven't met many engineers that didn't think, that given enough time, they could figure out and understand anything and everything.

sierra · Mar 6, 2011

Myself, I think Ive come up with reasonable plan for being safe and successfull in the mountains. Climb and hike for 20 to 30 years, averaging 75 to 150 climbs per year, do not let conditions or weather stop you, unless its borderline suicidal to continue. Fall, freeze, use inadequate gear, then upgrade said gear. Hike with sane and insane partners, male and female, make sure to get in over your head a least a few times a winter. Continue on when you shouldnt, then fail, when your borderline hypothermic, you can go home. At the end of your 20 to 30 years, you should be good to go.

Paradox · Mar 7, 2011

I personally would be very interested in keeping tabs on your research. When I went to some of the links I noticed that "death" is considered failure and not some other unfortunate end to the event. I can see where death of the participant presents an incontrovertable endpoint, but it seems to limit the amount of data we would be able to submit.

Also, as I read your paper on McAuley, I was struck by the difference personal motivators would have on an individual. For instance, McAuley had spent perhaps 25% of his life, in one way or another preparing for that single event. He was willing to explain away a malfunction in the bubble control system by saying he was going to be "super careful", because much of his life preparedness was invested in that singlular event.

While for me, I might have spent 0.1% of my life preparing for a big hike. Also, I have a lot of time and effort invested in a wife, children, and a business that supports them. Turning around when the batteries in my GPS are running low is no biggie for me.

Tom Rankin · Mar 7, 2011

Paradox said:
When I went to some of the links I noticed that "death" is considered failure and not some other unfortunate end to the event. I can see where death of the participant presents an incontrovertable endpoint, but it seems to limit the amount of data we would be able to submit.
...
Also, I have a lot of time and effort invested in a wife, children, and a business that supports them. Turning around when the batteries in my GPS are running low is no biggie for me.

There are lot of ways to die. I think we should be clear on this. Apparently some people consider 'he was identified by dental remains' to be different from 'his frozen corpse was discovered ... ', or 'his body was never recovered', etc., etc.

Maybe another turn around factor besides DEATH should be

'avoid being the subject of endless online discussion after your incident is published online'.

Little Rickie · Mar 7, 2011

Tom Rankin said:
Maybe another turn around factor besides DEATH should be

How could you quantify it accurately? The opinions and discussion on that would be endless.

Accident Prevention Using Failure Analysis

Help Support vftt.org:

New member

Well-known member

Active member

New member

Well-known member

New member

New member

New member

New member

New member

Well-known member

New member

Well-known member

New member

Well-known member

New member

Well-known member

New member

Well-known member

New member

Similar threads

Join the conversation!

Join today and get all the highlights of this community direct to your inbox. It's FREE!