Probability vs. Likelihood in OHSAS 18001 Clause 3.21

[mirrorcrax]

In OHSAS 18001:2007 it states under clause 3.21 "risk - Combination of the likelihood of an occurrence of a hazardous event ......"

but when i looked up likelihood in Google this came up:

"Likelihood is the hypothetical probability that an event that has already occurred would yield a specific outcome. The concept differs from that of a probability in that a probability refers to the occurrence of future events, while a likelihood refers to past events with known outcomes." on a site called wolfram mathworld

however when we're conducting a risk rating or a risk evaluation whatever you want to call it, likelihood represents the possibility of a specific occurring in the future

Now i'm confused, does it seem right to you, how can likelihood statistically mean that it is the probability of a specific outcome in past events, while in OHSAS18001 it is part of the risk evaluation of future events

can someone please shed some light on this

[Jennifer Kirley]

I'm tempted to believe you are overthinking this subject of risk.

I don't have this 2007 version. Can you please give us the rest of the clause? In addition, the clauses before and after would help.

[Randy]

Not that I know anything about 18001 or risk assessment, but I'll venture a guess.

1st let me say that you learned risk assessment as a child when you were taught to cross a street and have used the process many times a day, every day of your life.. When you were young hopefully you were told that when crossing a street to stop, look both ways, and when it was clear to cross. The advent of crosswalks and pedestrian signals brought forth physical control measures.

The instructions you received were a control measure containing criteria meeting 4.4.6.

The stopping and looking had you take the time to determine probability (possibility of getting run over, this is also a control measure) and the consequences of getting hit (seriously injured or killed). When cars are approaching and or close the probability is high, when they are absent they is no probability or far away the probability is non-existant or low. The consequences remained the same. And there you have it, the risk analysis process.

Just think of probability as "possibility" taking into consideration past history and existing control measures.

Don't go crazy with this and make it too complicated, it's as easy as crossing the street.

Here's a simple way to do it from Canada

[mirrorcrax]

Thanks so much for replying Jennifer and Randy,
and thanks Randy for the simply put and brilliant example,

Section 3 is "Terms and Definitions"

clause 3.21 states "risk - Combination of the likelihood of occurrence of a hazardous event or exposure(s) and the severity of injury or ill health (see 3.8) that can be caused by the event or exposure(s)"

I get the idea, and how to perform a risk assessment and all and how to identify hazards, but i have doubts as to the use of "likelihood" in the standard when what really matters is "probability", and when i re-deliver information i want to use the right term because i have seen them in presentations used interchangeably though statistically they have different meanings

[Randy]

Liklihood can be derived from your organizations history with events...

Look at what you've been doing with the already existing control measures whatever they are....Have you had any incidents?

If no, then the liklihood is low

If yes, you've had incidents, then the liklihood would be elevated and you need to put into place better controls like procedures, signage, training or whatever that can bring that liklihood to a more acceptable level, thereby reducing risk.

[Jim Wynne]

Originally Posted by mirrorcrax

=
I get the idea, and how to perform a risk assessment and all and how to identify hazards, but i have doubts as to the use of "likelihood" in the standard when what really matters is "probability", and when i re-deliver information i want to use the right term because i have seen them in presentations used interchangeably though statistically they have different meanings

Don't worry about the arcane difference in the definitions. The standard uses "likelihood" in its common-usage sense in which it's synonymous with "probability."

[Haneen]

Subjective part (personal judgment) is an integral part of risk assessment regardless its type (quantitative, semi-quantitative or qualitative RA), for example in the quantitative risk assessment which completely based on probabilities (some times called probabilistic RA) there is some subjectivity because one is using past/heroically data to predict future event that did not happened yet.

Likelihood is the subjective face of probability .. probability is presented as figure: 1 , 0.09; 0.0005 .. etc (it can't be more than 1). Likelihood is a word or even number to describe your judgment about specific issue (High, Medium, Low which can be 1,2,3 but it is still subjective/likelihood) - other classifications (risk matrix) using 5x5 scale instead of 3x3 scale that i used in my illustration.

Example:
Cable is placed across the staircase: likelihood of that someone may trip & fall is HIGH, if the cable is placed so close the wall then the likelihood of the risk of someone trip & fall, been realized will be LOW ... This is just a personal judgment (the quality of such personal judgments will be enhanced through work experience)

Regards

[Randy]

Thank you, but this Thread is 18 months old.

Additionally, the numbers you supplied are nice except in cases where the person determining risk is illiterate.

You are correct about the subjectiveness of risk assessment.