Rank: New forum user 
|
For reporting purposes, I need to set up an Excel sheet to automatically tally up Unsafe conditions, Acts, Near misses, first aids and LTIs and present them in a Heinrich pyramid format. I need some guidance on the methodology/logic of the Heinrich pyramid please. Any input welcomed. Here is my setup:
Unsafe Act/Conditions are added directly to the Heinrich tally.
Near Misses are investigated and root cause determines as UA or AC. This root cause is added to the tally for Heinrich. So 1 Near Miss generates 1 Near miss on Heinrich + 1 UA/UC on Heinrich.
First Aids are added directly to Heinrich. Then investigated to determine root cause as UA or UC and corrective actions taken. 1 First Aid generates 1 First-Aid on Heinrich + 1 UA/UC on Heinrich
Now for my problem:
When First aid goes to Lost-Time-Incidents (seperate levels for <3days, 3days<LTI<30 days, >30 days), should the tally cascade upwards, or should they replace each other.
In practical terms, if I have 10 first aids that are off for 35 days, would this mean:
a) On Heinrich: First-Aid=10, <3days=10, 3<LTI<30=10, >30=10
b) OnHeinrich: First-Aid=10, <3days=0, 3<LTI<30=0, >30=10
I'm looking at this from a reporting viewpoint only, logically i would assume (a), but to avoid extra "ghost" totals I would say (b). Any guidance...
|
|
|
|
|
|
Rank: Super forum user 
|
Brendan
All a bit too 'whizzy' for an old git like me. However, the Heinrich/Bird pyramid is a metaphor illustrating the hypothetical link between fatalities, major injuries, etc. There will be some correlation between the different categories and how you present is them is purely your choice. If using a pyramid I suspect it will be a bit lop sided.
|
|
|
|
|
|
Rank: Super forum user 
|
WOW!
I have the same problem :)
to try and keep it simples I record
near miss
Incidnet
RIDDOR
We then mathematically extrapalte up (kind of to try and calculate fatality rate but this gets difficult :) )
To present the 'pyramid' i make s a horizontal bar chart. the only way i have found to make it pyramid shaped is to use negative numbers by halving results and displaying accross a 0 axes... does that mak sense.
my only aim is to show our pyramid is top heavier to encourage reporting, and to show how serious the effects could be if we don't address..
I could share off forum if you PM but it's not the best effort ever :)
|
|
|
|
|
|
Rank: Super forum user 
|
Using the Hienrich pyramid to set out your figures depends on how you set the equivulent fatalities measure. I assume the fatality figure is a theretical number not actual fatalities. The rail industry used a set formular to measure this, but am not sure of the precise numbers or how they calculated them. Try and contact the RSSB and ask them how it is done. Thier website may have some very usefull info.
|
|
|
|
|
|
Rank: Super forum user
|
bob shillabeer wrote:I assume the fatality figure is a theoretical number not actual fatalities.
Try and contact the RSSB and ask them how it is done. Their website may have some very useful info.
:)
Yup very much theoretical and used a stick! It's also massively over predicted as we use our over 3 day in the major box.... The idea is to make people think about near miss, I should have made that clearer! :)
Thanks for the link!
|
|
|
|
|
|
Rank: Forum user
|
In a past life I used to do a lot of data analysis and from where I sit and putting myself as if i was being presented with the data I can say that either would be fine, so long as it is made clear on any representation of the data you make. This way the people who look at it, if not just for yourself, have an understanding of the way you work and your findings.
Once the model is set up you could run both formats on separate tabs/speadsheets with one populating the other automatically and decide which set of data works for you best.
If i had to chose one i would have gone for the option where no repetition of data occurs (ie no ghosts). If you have a broken leg and a bruised arm as a result of the same incidnet do it go as a major injury only or as a major injury and a minor injury, same principle as the initial question.
As a matter of interest we currently use a third party to categorise all our incidents and they go for the simple version mentioned earlier in the post. (i think about 5 different categories)
|
|
|
|
|
|
Rank: New forum user
|
Thanks Anthony. This answers my question nicely. I'll be going with the no-ghost option.
|
|
|
|
|
|
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.