If you require further information on any of our services please do not hesitate to contact us by telephone or e-mail. All enquiries are extremely welcome.
T: 01636 816 466
The Old Vicarage
Station Road, Rolleston
You can always be more efficient
Operational Research is all about using analysis to improve decision making. Dr Angela Moore uses mathematical modeling, statistical analysis, optimisation, sequencing and scheduling techniques to create simple heuristics you can employ to run your operation more efficiently.
Wednesday, 09 May 2012 14:47
Now everyone knows about sickness rates, where the average is 2.5-3%. But the real question is what this sickness rate means to the operation? Well this is a very complicated question. Firstly you have to consider how the sickness rate is created. Normally it is calculated by the number of sickness hours per year divided by the contracted hours. Well this is a very useful statistic if you want to know how many hours a group of staff will be sick for each year. But it does include both long term and short term sickness as well as sickness at work and on holidays. So when it comes to arranging cover for these absences, you may not treat all of these cases the same. For instance if someone is going to have an operation, it will normally be booked a long time in advance. And you would know if they were likely to be back at work soon after the operation or if they would need recovery time. So you would not arrange cover for this type of absence in the same way you would deal with covering for a shift where someone fell ill at work.
Secondly, the danger with statistics is that they need to be used within a context. So at a university for example you will have a large range of diverse staff. Now if the university just has one sickness rate this can mislead you when organising cover for a specific department. After all the lecturers may have an average absence rate because they are mixing with students all the time and catching every bug, yet the job is low risk. The administration may have a very low sickness rate because they have a risk free job. The security department may have a high absence rate because of the risk involved with the work. So you need to know the absence rate for the department if possible.
Thirdly, to minimise the disruption to the operation you would use this average statistic to create a cover arrangement. So if the statistic says that you have a 5% sickness rate how many times will one person be absent off a shift. How many people will need to be on cover per shift to ensure you will always have the correct number on duty? How any people to ensure that 90% of the time?
Well the answer is not as straight forward as you would think. Let’s say the operation is 24/7, 12-hour shifts with 14 on duty per shift. The graph shows the number of people that will be off a percentage of shifts for a 5% absence rate. With 12 hour shifts, 24/7 there would be 730 shifts per year. So on less than 50% of the shifts no one will be absent. You would expect one person to be absent on 36% of the shifts, that’s 252 shifts per year. You could have up to four people absent on any shift, and you would expect that to happen three times per year. You would need three people on cover every day to ensure that 90% of the time you had the correct minimum staffing numbers. That’s will cost 11% extra in labour charges for 5% absence. Three extra would give you the correct minimum staffing levels on 95% days. This would still leave you short on 18 days per year.
This sort of analysis is vital for ensuring you have the correct number of staff when and where you want them. We do this analysis as part of our Business Health Check, and whenever we set up a new shift pattern.