You are not logged in.
Pages: 1
Bob,
Thank you so much for your time and your thoughts. It is always good to bounce your ideas about the place as it makes you re-evaluate the original problem. This is coloquially know as the KISS method (Keep it Simple Stupid).
Thanks again,
Seals
Bob, that data was collected over the year of 2010 in a remote area of Australia. So the numbers are not high. Just for my pure amusement could I use the weighted average to forecast staffing requirements for new year (I would have the forecasted number of jobs) or would it be easier to follow the solution that you have provided? Being that 1682 hours would be required to complete all jobs, which as you have pointed out would leave the person with time to spare (which coincidently is what I was expecting as an outcome)
I think that I may have over thought the issue at the onset?
Yes that is right but job C is a quick job that takes say 1 hour where job B takes on average 3 hours. So you are right we need to find out how long it takes to efficiently complete all the jobs over one year. The average is useful to forecast what would be likely to occur next year.
So we can forecast that the jobs will decrease by say 5% next year we can then with the average forecast how many staff will be required to complete the tasks. So really it is not only historical but future orientated based on outcomes (if that makes sense)
The time period is one year, the weights are based on the amount of time it takes to complete each task (This part will take time to refine etc)
ok staying on line. fyi the incidents are actually events that have taken place and i am trying to work out how many people it would take to efficiently complete those tasks. On average a worker would spend 8 hrs a day, 38 hours a week for 45 weeks on average per year. this figure is where I get the 1710 from
Bob,
ok getting clearer now Thankyou for your time and patience
So now we have a weighted average of 4.85 to work out the time correlation is it just a matter of multiplying the 4.85 by the number of occurrences which would be 1682.95? I am taking this to be the amount of hours it would take to (on weighted average) complete the tasks over a year?
Hi Bob,
Thanks for that, I understand your formula and the theory thereof. If I divide 22.10 by 347 I get an extraordinarily small answer 0.0637, however, if I divide 347 by 22.10 I get 15.7 which seems more believable I think.
I have one more part to this. When I get my weighted value by the supplied formula I need to correlate that to the amount of hours worked in one year. In this case that amount is 1710. Is it simply a case of multiplying the weighted average answer by the number of occurrences?
Thank you very much for your help thus far.
Seals
Hi experts,
I am looking for some help with a weighted average function. I am trying to work out the average time taken per incident, but I wish to weight or grade the incident types, The data:
Incident Occurences Weight Weighted Value
Incident A 1 45% 0.45
Incident B 72 5% 3.60
Incident C 138 5% 6.90
Incident D 49 5% 2.45
Incident E 11 10% 1.10
Incident F 15 10% 1.50
Incident G 56 10% 5.60
Incident H 5 10% 0.50
My calculations are as follows, obviously 45% of 1 gives 0.45 as a weighted total, all of the weight %'s sum to 100%, all of the weight values sum to 22.10. I am currently dividing the weight total (22.10 by 8) as there are 8 incident types? This gives me 2.76 this is then multiplied by the total number of incidents (347) which gives me a weighted total in hours of 959 hrs.
Am I correct in my assumptions/calculations or am I looking at this the wrong way.
Any help would be greatly appreciated:cool:
Cheers,
Seals
Pages: 1