Posted By Bill Elliott
Three slightly different theories -
Heinrich (1931) - for EACH major accident there are likely to be 29 minor injuries and 300 non injury incidents.
Bird (1969) - for EACH serious injury there are likely to be 10 minor injuries, 30 property damage accidents and 600 non injury accidents. To add further to this
Tye/Pearson (1975) - for EACH fatal or serious injury there would be 3 minor injuries with absence of up to 3 days, 50 injuries requiring first aid, 80 property damage and 400 non injury/damage accidents.

Posted By fats van den raad
Lewis
These figures are not just plucked out of thin air or derived from someone's gut-feeling, but are actually the results of research. As far as Bird's triangle goes, my own research some time ago when I was involved in behavioural safety showed a significant correlation to the figures in Bird's triangle.

Yes, I'll give you that a thousand paper cuts are not likely to manifest in a fatality (unless it's the same person maybe??) But the chances of that happening is as great a statistical possibility as having only a thousand paper cuts a year, and nothing else.

Posted By Merv Newman
There has been an awful lot of discussion over the years on these triangles. The main hang-up is usually on the definition of terms.

Though not 100% convinced myself I usually quote Byrd's 1 : 29 : 300 even though it was developed for american industry.

I do know of one city ring road which I use frequently for which the stats have been consistently in the range of 10 fatalities per year (+ or - 2) and 300 (+ or - 20)serious injuries. Nobody counts the equivalent of first aid injuries ie bent metal and broken glass.

The installation of some fixed speed cameras early this year has reduced the actual numbers by about 25% but the ratios may well stay the same.

And, if you are having a 1000 paper cuts a year then it is statistically probable that the quality of your safety management system is such that a fatal injury is quite likely. (spoken like a stochastical man)

Posted By Lewis T Roberts
Fats,

I have already assumed that the figures are not plucked out of the air or gut feeling, I was just commenting on the validity of the relationship between the top and bottom of the triangles given that there is overwhelming evidence of under reporting of over three day and below type accidents and given that there are countless industrial sectors with differing accident incident rates (AIR).

Lew

Posted By fats van den raad
Lewis
I hear what you are saying about under reporting and of course, this would have an impact on any such ratio's. The research I have done myself was in an environment wher I had considerable amount of confidence in the reporting both in numbers and in accuracy.
I have actually heard a school of thought that argues that if your own accident stats do not follow the trend indicated by the triangles, i.e. that there are less incidents on the bottom tiers of the triangle, this in itself indicates under reporting. I would not know if I would go as far as that in my faith in the ratios....

I tend to look at the general shape rather than get too hung up on actual numbers. Fact is the more minor injury accidents you have, the more chance you have of having to deal with a more serious accident.
If you control the detail that is required to eliminate the minor injury accidents, most, if not all, of the issues involved in the more serious accidents will be controlled by default.

Posted By Merv Newman
Vincent,

"questioning" the validity of the triangles, but not necessarily disproving it. That a statistical pyramid exists is perfectly logical. We all know that first aid injuries occur more often than fatalities. (thank your preferred religious figurehead) It is the actual shape of the pyramid that is in question. Pick any one of the quoted "pyramids", make sure you understand and agree with the definitions, and go with it.

I have seen a report (where ?) that when heavy incentives are provided for not taking time off, then lost time injuries go down but the total number of injuries stays constant. This indicates that "cultural factors" can influence the shape of your triangle. And the value of defining and measuring "recordable" injuries. ie, anything that REQUIRES the attention of a medically qualified practitioner (doctor)

I think the "Recordable Injury Rate" is a good overall measure of how many times your safety system is failing it's customers.