FireNet Community
FIRE SERVICE AND GENERAL FIRE SAFETY TOPICS => Technical Advice => Topic started by: Benzerari on November 14, 2008, 08:48:28 PM
-
Say you have done a fire risk assessment, for a given building, to what extent, can you tell the fire will not either kill or damage... in case fire will occur?
I mean in terms of percentage!
-
What I mean by that is; say some one may say, this can never be applied to fire safety. What about medical field? If a doctor says there is 70% chance a patient can either survive or dye... etc, why not the same principle mayn't be applied to fire safety? when saying there is 70% fire can or can not strike in this building...!
-
Easy. There is a 99% chance you are won't be injured from fire in any building. Before and after, with or without a risk assessment.
-
Easy. There is a 99% chance you are won't be injured from fire in any building. Before and after, with or without a risk assessment.
In what basis, did you get 99% chance of none injury?
-
I was once told the way to get to that sort of figure was to build out of stainless steel and don't let anyone move in.
-
I've actually been sent plan drawings submitted by artictects which have consisted of rooms with no door openings indicated on the plans.
Very safe as people cannot enter/exit the rooms
Just an ideal world!!!!!!!!!
-
I've actually been sent plan drawings submitted by artictects which have consisted of rooms with no door openings indicated on the plans.
Very safe as people cannot enter/exit the rooms
Just an ideal world!!!!!!!!!
Even that wouldn't be safe enough, fire can strike from out side rather than from inside, just as an example: Annually in the U.S. there are more than 300,000 fires that originate in homes. In addition, nearly 10 percent of the land and over
one-third (42 million) of the homes in the U.S. today belong to the Wildland Urban Interface (WUI). The WUI is used to refer to both areas where housing abuts heavily vegetated areas (interface) and those areas where houses and vegetation are intermingled (intermix). If current trends in housing continue, the WUI will grow rapidly.
By means fire may strike in any building and spread out to others... see the link below..., would FRA take into account that?
http://news.yahoo.com/s/ap/20081115/ap_on_re_us/wildfires
-
An FRA will never make a premises safe. The outcomes of the FRA may assist but there is always the human element to consider.
-
An FRA will never make a premises safe. The outcomes of the FRA may assist but there is always the human element to consider.
What I mean is; to what extent (percentage), it could or couldn't be safe? I agree 100% safety, just doesn't exist :)
It seems probability and statistics don't apply to fire safety in that way, except from statistics figures that are compiled after fire events, and the question is still why? So many disciplines applying probability and statistics, why still aren’t they applicable to fire safety, to predict safety?
-
Because there are too many vairables and because risk = liklihood + severity.
-
Because there are too many vairables and because risk = liklihood + severity.
That's what probability and statistics are for, even rigorousness is quantified with certain percentage ..., any business’s management and marketing risks can be quantified and/or predicted..., and so many disciplines dealing with so many variables and some with more than two degrees of variations... :)
-
Because there are too many vairables and because risk = liklihood + severity.
That's what probability and statistics are for, even rigorousness is quantified with certain percentage ..., any business’s management and marketing risks can be quantified and/or predicted..., and so many disciplines dealing with so many variables and some with more than two degrees of variations... :)
I disagree. Business is more complex than that and I don't think it is possible to put a numerical figure to everything. Show me some examples of the percentages for the quality of the business management for your company, FireNet or some famous PLCs. Life has many variables which are unknown.
-
Because fires are caused by human beings and because of the complexities and unpredictable and sometimes irrational behaviour of this animal one can never quantify the risk they pose to themselves and others.
It is probably easier to identify a pattern to falling raindrops.
-
Because there are too many vairables and because risk = liklihood + severity.
That's what probability and statistics are for, even rigorousness is quantified with certain percentage ..., any business’s management and marketing risks can be quantified and/or predicted..., and so many disciplines dealing with so many variables and some with more than two degrees of variations... :)
I disagree. Business is more complex than that and I don't think it is possible to put a numerical figure to everything. Show me some examples of the percentages for the quality of the business management for your company, FireNet or some famous PLCs. Life has many variables which are unknown.
In any business, if the managers don't predict that more than 50% the business will be profitable, they wouldn't setup the business :)
-
Say you have done a fire risk assessment, for a given building, to what extent, can you tell the fire will not either kill or damage... in case fire will occurs?
I mean in terms of percentage!
Never. The most precise 'scientific' forms of risk assessment (e.g. Quantified Risk Assement) are not necessarily accurate.
The UK fire statistics for 2006 (the latest that are available) State that in 2006 there were 32,900 fires recorded in buildings other than dwellings. Of these 21% were in private garages and sheds (buildings that don't require a fire risk assessment), so that leaves roughly 26,000 in buildings likely to be covered by the Fire Safety Order. In the same period 37 people died in fires in buildings other than dwellings (i.e. on average 0.14% of fires in relevant non-domestic premises lead to death).
The regulatory impact assessment for the RR(FS)O admitted that it was difficult to predict the number of premises that the Order applied to in England & Wales, but estimated it as 2,153,140.
So... based on the 2006 statistics the average chances of having a fire in a premises in England & Wales coverd by the RR(FS)O is (conservatively, because the number of fires includes Scotland & N. Ireland) - roughly a 1% chance per year. The average chances of anyone dying in a fire if you've got one of those premises is roughly 0.002% per premises per year. Of course, risk doesn't even out like that, but the fact is that it is an extraordinarily unlikely probability!!!
-
You should take in to account the percentage of fires that are reported to the fire service. Most are not.
-
The probability of fire occurrence ‘Pf’ should be function of so many variables including:
1 - 'P1' percentage related to previous statistics and it should be at least multiplied by a coefficient of 1.5, since the real world statistics are far more serious than the one recorded and compiled by statisticians... etc
2 - 'P2' the percentage of availability of combustible material in the building compared to other none combustible ones... etc
3 - 'P3' the percentage related to the medium temperature within the building deducted from a year time records... etc
Probability of fire occurence ‘Pf’ = f (P1, P2, P3, ...) = P1 + P2 + P3 + ... etc
In addition to other variables where applicable... etc, also each sub probability could be function of others sub-variables… etc
What do you think guys?
-
The probability of fire occurrence ‘Pf’ should be function of so many variables including:
1 - 'P1' percentage related to previous statistics and it should be at least multiplied by a coefficient of 1.5, since the real world statistics are far more serious than the one recorded and compiled by statisticians... etc
2 - 'P2' the percentage of availability of combustible material in the building compared to other none combustible ones... etc
3 - 'P3' the percentage related to the medium temperature within the building deducted from a year time records... etc
Probability of fire occurence ‘Pf’ = f (P1, P2, P3, ...) = P1 + P2 + P3 + ... etc
In addition to other variables where applicable... etc, also each sub probability could be function of others sub-variables… etc
What do you think guys?
Building alteration may affect the the probability of fire occurence 'Pf' therefore It has to be taken into account...
What do you think?
-
The probability of fire occurrence ‘Pf’ should be function of so many variables including:
1 - 'P1' percentage related to previous statistics and it should be at least multiplied by a coefficient of 1.5, since the real world statistics are far more serious than the one recorded and compiled by statisticians... etc
2 - 'P2' the percentage of availability of combustible material in the building compared to other none combustible ones... etc
3 - 'P3' the percentage related to the medium temperature within the building deducted from a year time records... etc
Probability of fire occurence ‘Pf’ = f (P1, P2, P3, ...) = P1 + P2 + P3 + ... etc
In addition to other variables where applicable... etc, also each sub probability could be function of others sub-variables… etc
What do you think guys?
Building alteration may affect the the probability of fire occurence 'Pf' therefore It has to be taken into account...
What do you think?
Should you not include the human factor?
-
Why complicate things and try to quantify things that cannot be measured.
What is the probability that I will be knocked down and killed by a yellow ford focus whilst crossing the road outside 60 Old Kent Road at 1032 pm tonight having drunk two pints of Fullers? What if I drink three pints of Stella?
Answer - absolutely zero. I only drink wine.
KISS
-
Why complicate things and try to quantify things that cannot be measured.
What is the probability that I will be knocked down and killed by a yellow ford focus whilst crossing the road outside 60 Old Kent Road at 1032 pm tonight having drunk two pints of Fullers? What if I drink three pints of Stella?
Answer - absolutely zero. I only drink wine.
KISS
Kurnal;
I think every thing can be quantified at least with probabilities... etc, it's rather our understanding and analysis towards seeing things, which haven't yet achieved to decipher the necessary know how, to do so... etc
-
But probabilities are always approximations and always include error, the more variables the greater the margin of error and once you start multiplying error by error the result is so inaccurate it becomes dangerous. And besides human factors - failure to behave in the expected manner - is at the root of most fires. This failure may cause a fire through carelessnes or intent, the failure to install or maintain properly may cause the consequences of the fire to be more serious than they should be, the persons may behave in the wrong way and this may affect their ability to make a safe escape.
Why dont you have a read of BS7974 which approaches fire safety from an engineering basis and the interactions of sub systems affecting fire safety- it may be what you are looking for.
-
The probability of fire occurrence ‘Pf’ should be function of so many variables including:
1 - 'P1' percentage related to previous statistics and it should be at least multiplied by a coefficient of 1.5, since the real world statistics are far more serious than the one recorded and compiled by statisticians... etc
2 - 'P2' the percentage of availability of combustible material in the building compared to other none combustible ones... etc
3 - 'P3' the percentage related to the medium temperature within the building deducted from a year time records... etc
Probability of fire occurence ‘Pf’ = f (P1, P2, P3, ...) = P1 + P2 + P3 + ... etc
In addition to other variables where applicable... etc, also each sub probability could be function of others sub-variables… etc
What do you think guys?
Building alteration may affect the the probability of fire occurence 'Pf' therefore It has to be taken into account...
What do you think?
Should you not include the human factor?
Just nearlythere good point :) I mean it's a good point
-
Why dont you have a read of BS7974 which approaches fire safety from an engineering basis and the interactions of sub systems affecting fire safety- it may be what you are looking for.
Have you got a copy of BS7984 please
-
Yes.
-
Your local library will be able to help you. BS7974 is copyright and runs to several hundred pages.
-
But probabilities are always approximations and always include error, the more variables the greater the margin of error and once you start multiplying error by error the result is so inaccurate it becomes dangerous. And besides human factors - failure to behave in the expected manner - is at the root of most fires. This failure may cause a fire through carelessnes or intent, the failure to install or maintain properly may cause the consequences of the fire to be more serious than they should be, the persons may behave in the wrong way and this may affect their ability to make a safe escape.
Kurnal;
What I meant from this topic is that, one day the fire risk assessment may achieve some ways to end up by saying, the building will be 70% safe by means there are 30% chance fire can strike..., it's all about predictions in more scientific and engineering ways, probabilities have been used for so long time for predictions processes, the weather forecast is just an example of that, when saying tomorrow it will be shower in the morning till mid day... and sunshine till the end of day with clear sky...etc, even their predictions fails some times. At the end of the day, they are just predictions based on probabilities, statistics and mathematical calculations, while taking into account so many variables and parameters of both major and minor effects..., even though some times the prediction itself revealed to be far away from the reality..., therefore it's this failure in calculation that must be taken into account while investigating, in order to either correct, or find out other substantial directives and solutions in predicting the probability of fire occurrence...etc
-
But probabilities are always approximations and always include error, the more variables the greater the margin of error and once you start multiplying error by error the result is so inaccurate it becomes dangerous. And besides human factors - failure to behave in the expected manner - is at the root of most fires. This failure may cause a fire through carelessnes or intent, the failure to install or maintain properly may cause the consequences of the fire to be more serious than they should be, the persons may behave in the wrong way and this may affect their ability to make a safe escape.
Kurnal;
What I meant from this topic is that, one day the fire risk assessment may achieve some ways to end up by saying, the building will be 70% safe by means there are 30% chance fire can strike..., it's all about predictions in more scientific and engineering ways, probabilities have been used for so long time for predictions processes, the weather forecast is just an example of that, when saying tomorrow it will be shower in the morning till mid day... and sunshine till the end of day with clear sky...etc, even their predictions fails some times. At the end of the day, they are just predictions based on probabilities, statistics and mathematical calculations, while taking into account so many variables and parameters of both major and minor effects..., even though some times the prediction itself revealed to be far away from the reality..., therefore it's this failure in calculation that must be taken into account while investigating, in order to either correct, or find out other substantial directives and solutions in predicting the probability of fire occurrence...etc
What do you think about this?
-
Kurnal;
What I meant from this topic is that, one day the fire risk assessment may achieve some ways to end up by saying, the building will be 70% safe by means there are 30% chance fire can strike.....
The liklihood od a fire occuring in any year is less than 1% in more then 99% of buildings.
-
Kurnal;
What I meant from this topic is that, one day the fire risk assessment may achieve some ways to end up by saying, the building will be 70% safe by means there are 30% chance fire can strike.....
The liklihood od a fire occuring in any year is less than 1% in more then 99% of buildings.
How did you calculate that?
-
In my head. But it is true. All you need to know is how many fires there are and how many buildings there are in the UK.
-
Is that all you need to know?
-
I think I know enough to reach that conclusion. Others are free to have different opinions and offer them. That's how FireNet works. :)
-
I guess there are more to take into account not just the previous fire statistics and the number of buildings, for any given building it has to have a special case study to know what's what... etc :)
-
Benz
You will find an article in the February "Fire Risk Management" in which Michale Belcham as given an overview and worked example of a probabilistic fire risk assessment. Its by far the best article on the subject I have seen- even if I dont understand it ;).
-
Benz
You will find an article in the February "Fire Risk Management" in which Michale Belcham as given an overview and worked example of a probabilistic fire risk assessment. Its by far the best article on the subject I have seen- even if I dont understand it ;).
If you don't mind.... ;)
-
Benz,
Read your article in FSE, Very interesting....
P.S. sorry for going off thread....
-
Benz,
Read your article in FSE, Very interesting....
P.S. sorry for going off thread....
It's alright, thank you Ark Angel and welcome to all comments and suggestions :)
-
Benz you have a message.
-
Benz,
Read your article in FSE, Very interesting....
P.S. sorry for going off thread....
It's alright, thank you Ark Angel and welcome to all comments and suggestions :)
must of missed that and i had the magazine. Do you have it on file benz?
-
Benz,
Read your article in FSE, Very interesting....
P.S. sorry for going off thread....
It's alright, thank you Ark Angel and welcome to all comments and suggestions :)
must of missed that and i had the magazine. Do you have it on file benz?
It's on FSE online, and it's the same article posted in the forum a while ago just reworded a little bit.
Click: http://www.fseonline.co.uk/articles.asp?article_id=8429&viewcomment=1
-
strewth
-
Benz you have a message.
Thanks Kurnal, you have one too.
-
This is the first time, I have heard that PRA (Probabilistic Risk Analysis) is part of BS7974, but unfortunately still under-utilized by fire safety engineering, the article of 'Belsham' is deeply focused on the original subject of this thread...,
In one side this is a good confirmation that fire safety as an engineering discipline is far yet to be an exact science, similarly to other catastroph disciplines such as 'Seismiology', 'Volcanology'..., or even 'Weather Forcasting Engineering'...
-
Interesting reading in the main subject
http://www.fpemag.com/articles/article.asp?i=402