We are frequently called in to work with companies to develop plans for various scenarios- new product introduction, competitor response, disaster prevention, among them. The motivating factor is to maximize results and minimize risks. Those platitudes are what everyone wants- but most of our clients have not a clue how one must analyze risks.
There is no “risk-o-meter”. No device that we can plop into a situation and say with certainty that there is a 23.4% probability of failure. No system that would have assessed with accuracy the damage that Hurricane Irene would create with flooding in Vermont and not with wind damage to New York. Or, when the earthquake hit DC, what risk our federal “safety officers” were promoting with their employees and visitors by demanding the evacuation of their high rises and standing in the street between them for two hours.
(Come on- if a building were to fall- any building among the nine contiguous ones where I was forced to wait- the odds clearly favored my destruction on the street, rather than in one of nine buildings! So much for risk management.)
I remember being told by Alfred Kahn, then the Director of the Civil Aeronautics Board, that they expected one passenger incident for every X million passenger miles flown. (I honestly can’t recall the actual number- but back then, it was 9 or less.) So, my accident was clearly expected, since I had flown about 800,000 miles a year for the past ten years. For a smart guy, that was a perplexing utterance.
There is a vast difference between one fatality every 9 million miles flown and ten airplane crashes with 100 fatalities over the course of a year. Because it’s the number of take-offs and landings (which is when most accidents occur) that set that number- not my 150 round-trip, cross-country flights. To evaluate a risk, we need to determine the contributing components.
There also is a difference between the frequency and the severity of a risk. If there is 1 fender bender every day on a certain five mile stretch of a road (for 365 accidents a year); is that the same risk as another five mile stretch that has a bus-car crash every week, with five or more fatalities each time? Especially if 5000 cars travel the first road daily and 10 buses and 500 cars travel the second.
You see risk is the product of frequency(ƒ) and severity(C), the probability something will happen and the consequences when it does. In math…
RISK = ƒ X C
A low severity, high frequency event is one that has significant data present with which one can determine root causes and develop preventative measures. But, a low frequency, high severity event (like an aircraft hitting the World Trade Towers) provides very little data with which one can effect preventative measures in building structures.
To be honest, that is exactly the problem faced by Homeland Security when they consider issuing an alert. And, why our firm- which is located in the Dominion Power black hole zone (where every storm- except the last one- causes a 1 day to 9 day power outage)- has a disaster plan to satisfy our client needs, when we are left stranded. Our clients can’t be.
Roy, I am curious about your accident – when, how and what happened? I’m so glad that you live to tell about it. Interesting post about risk measurement. We all want to minimize our risks and I run the risk of overprotecting my children in trying to keep them safe. It’s a tricky balance sometimes! ~Suerae
Oh, Suerae, there was more than one! And, I was involved in risk assessment long before I had that opportunity.
The funniest (as in, now that I am not on that plane) was when my Piedmont flight had an engine fail during flight (with the requisite puff of black smoke) and the pilot said something to the effect of “Don’t worry folks, this happens all the time. We’ll get you home safe…” I know he meant it as reassuring, but I was much less sanguine.
Roy
What? More than one? ugh! Well, I guess the pilot was right – he did get you home safe. Were you injured and exactly how many accidents? You must be like a cat with nine lives.
Actually, my friends swear each of my “cats” has nine lives- and there are nine cats…
Roy
Numbers and probabilities. I thought there was a risk matrix for most situations, but of course not possible in real life. Lucky for you Roy you have nine cats 🙂
Maureen:
There could be a risk matrix for any situation- if you want to develop one! But, no, there’s no book that has a list that we can just pluck out the one we want/need.
Thanks for the comment!
Roy
Ok, besides how interesting the post is; I wanted to know something. Is that formula quite applicable for the determining of severity of frequency.
Sometimes we do enter a situation knowing the risks involved; like say driving a car way too fast; or maybe getting on to a roller coaster (though they are equipped with all safety, they have been accidents); the risk is high and the severity is high; but it does not mean that the frequency be low; I might tend to get on a roller coaster even if I have read about a thousand accidents and I know the risks involved. Or is that because I have been safe, I feel it is low risk? Do you understand what I am trying to ask or do I sound confusing? 🙂
Hajra recently posted..Will they call you over for a bloggers party?
I think so, Hajra…
I think you are saying many folks fly on an airplane. Knowing the risk of survival from a crash is low. (High risk.) And, assuming the frequency is low (not frequent). This makes the overall rate about 1 in 11 million. But, 3.9 people fly on planes each day- which means that there will be 10 people killed every month due to plane crashes. Lo and behold- that is what we see in the news.
The risk of dying in a car crash is pretty low. But the number of people driving each die is humongous. So, the overall risk is about 1 in 6000- or about 1000 times more likely than dying in a plane crash…
Is that it?
Absolutely!
Now, I am thinking twice before the hit the accelerator even on empty roads…
Hajra recently posted..Will they call you over for a bloggers party?
You can’t hear me laughing all the way to the Middle East!
Roy, I love this. I have been working on the study of risk management as opposed to Problem Analysis (Kepner-Tregoe). This article shows me that these are two separate systems. Thanks.
Ann Mullen recently posted..Developing a Quality Management System
Wow- K-T! That was a big program at Corning! I am so glad you found it of use, Ann.
You made my holiday weekend!