Posted By Kelvin George
Chris

You put your working day at 11 hours. Does this not intimate that the working day for each employee is then 11 hours.

In my last profession we had a working day of 24 hours however each employee only worked 12 of them and so the calculation was then based on 12 hour days.

Perhaps you should target the hours/day at what each employee would normally work - just a thought.

Cheers Kelvin

Posted By Sean Fraser
Chris,

I think you are mostly there already. Remember, it is only an average you actually need to apply in terms of worked hours, not an absolute.

You may want to classify certain positoons and work out the proportions of overtime they work, rather than apply a general average across the board, if this fits better, especially if there is a distinct disparity between those working in the higher risk activities against those working in lower risk activities. For example, office (adminsitrative) staff may only work overtime infrequently, but warehouse staff may seem to be working overtime frequently - and of course they will be working in a more hazardous environment and therefore more "exposed" to risk. Within this identified category, there may be some individuals who work no overtime at all and others who work a lot (Working Time aside!). Take a representative period, say three months, add all overtime hours together and average by number of individuals (absolute or mean, as desired) - that will be your tag-on to the standard hours per individual in that category.

To demonstrate the sensitivity of the calculation, simply change the overtime averages by double but keep all other values the same - see if there is any real change to the ratio. If it is negligible, don't worry about it any more! You can easily prove to detractors that the actual numbers worked are no improvement on the general averages you have have provided. As long as these are reasonably realistic, you should be OK. From what I can see it looks like you've made a good enough effort of thinking about it already, so just try it out.

Only one caution - make sure that whatever you decide and determine, it remains consistently applied. If you are frequently changing the calculation of the denominator, then the results produced will immediately become meaningless as there will be no anchor point and hence no comparative value. There should be only two variables - the number of LTIs and the number of personnel. In the example I gave there may be a difference in the number of employess within a category, and hence the number of exposure hours used in the calculation, but this is just a distillation of the same principle. Changing the overtime exposure should only really be done if there is a significant change to the average (e.g. the average per person moves from 2 to 10 - not 2 to 3!). After time, a gradual shift would need to be incorporated but only if it moves significantly away from the original baseline AND this is clearly noted in the next sequence of results, as you will have effectively set a new baseline.

Hope that made some sense - I'm not sure if I've managed to explain my thinking very well!

Posted By Ian Waldram
Whilst agreeing with the calculation methods given, my question is - why would you want to express the total as 'hours worked'? If the rest of the organisation measures things that way, they will have a 'feel' for what the numbers you publish mean. However, if they don't (and your initial question suggests that counting hours within your organisation isn't normal), then won't your new statistic just be a meaningless number?

Why not use a number the organisation already generates (e.g. a production statistic) and relate your last injury to that? Alternatively, why not just count days? The IOSH guidance on public reporting suggests that events per year per 100 people is a way of expressing data to which most groups can relate, i.e. they know what an organisation with 100 people 'looks like'. In round numbers, 100 people work about 200,000 hours a year, but as you've found no one counts hours today as they used to when time clocks were normal and the majority of employees were paid by the hour. So why do H&S people still use hours as a measure when the rest of the organisation no longer does?