In our case, a position refers to the hiring of one employee with specific constraints. For example, someone is hired to be cashier on a full time basis. Therefore, a schedule will be built using multiple positions (each with the same or different constraints). Positions are also referred to as a line or a row in a schedule. Each position will accept one or more shifts on different days in order for that position to be handed off to an employee.
That position also has a maximum capacity of work each day and each week. That position is also entitled to time off and sickness which reduces the capacity of the position. Absenteeism and rest can make an employee work only 75% of the time (depending on the number of vacation days, culture, break times, training, etc).
This means that even though you would have 40 hours of shifts to give every week, one employee is not sufficient. The number of positions you need are based on many factors:
- The number of shifts and their total durations: calculate the number of positions on the sum of shifts and not the workload
- The daily variation of shifts: There may be a different number of shifts required on each day for different positions
- The seasonal variation of workload: vacation resorts are exposed to seasonal demand where the number of positions will vary greatly during the year
- The demographics around your place of business: The people that surround you place of business make the pool of potential employees. If you are surrounded by students, don’t open full time Monday to Friday jobs or at least minimize them.
Calculating positions is sometimes seen as a chicken or egg story: should you create a schedule first and adjust the positions based on the schedule results, or should you estimate your positions and use them as a constraint on your schedule?
There are ways to calculate the positions you would require best to match your workload but they do ignore what you already have as employees. The important point in this step is that you must use the shifts that were built in step one and not use the sum of the workload as approximate numbers. The workload profile may come back to haunt you later down the road.
 Author’s irrelevant note: the chicken or egg story has been sorted out by Professor John Brookfield from the University of Nottingham who discovered that any animal’s DNA does not change during its lifetime which means that the egg had to come first with the DNA of the chicken.