Explains how labor, travel and other costs can be distributed over the duration of the resource or task
Labor Efforts vs. Time
When you plan for a resource to do a task you typically enter the total effort over the duration of that task, for example 1000 hours over a six month task. This could be one person full-time working the same number of hours each day (linear distribution) or several people up-front doing the bulk of the work in the first two months with one person part-time for the last four month (up-front distribution). It could be a gradual ramping up of resources over the task, peaking approximately half way through the task and then tailing off (bell-curve distribution). These different distribution options are shown graphically below.
Figure 1 - Bell Curve Distribution
Figure 2 - Linear Distribution
Figure 3 - Front Curved Distribution
Figure 4 - Back Curved Distribution
Figure 5 - Stepped Distribution
Which Distribution Curve Should I Use?
If you do not select a distribution curve then a linear or even distribution is assumed. This makes sense in the majority of cases, especially when you are not sure how the labor should be distributed over the duration of that task or WBS element. It is also the simplest and easiest to check the costs and hours in the costing model - since the hours and resource each calendar month will be directly proportional to the number of working days in that month overlapping the task or WBS-resource assignment start and end dates. "Full-time equivalent" (FTE) calculation of effort as task duration x FTE is also based on a linear distribution assumption.
- You can calculate effort from "full-time equivalents" and distribute that effort over the task using a non-linear distribution curve
For larger tasks having multiple resources assigned and having significant costs, it may be better to select a distribution profile best suited to the nature of the task you are planning. For example many projects with multiple resources will ramp up at the start - as people are recruited or on-boarded and the exact plan is being ironed out - and then ramp down at the end, as resources are rolled off. A stepped distribution curve suits this profile nicely, or a bell curve if the ramp up and ramp down is so gradual the resource load is only briefly peaks at the mid-point in time.
How are the Hours (or Costs) per Month Calculated from a Distribution Curve?
Labor hours or other direct costs are distributed between the WBS, task or assigned resource start and end dates based on the curve selected above. The costing model which is generated has a granularity of a calendar month, in other words on figure of hours and costs per calendar month. So how are the hours (or costs) distributed given that tasks can start in the middle of a month and do not necessarily start on the 1st and end on the last day of the month? The actual distribution works as follows - example data based on 1000 hrs of labor assigned to a task starting 1/15/19 and ending 3/25/19 and using a stepped distribution curve of 10%, 40%, 40%, 10% is also included:
- The total number of normal working days between the start and end dates, excluding holidays and weekends, is calculated. In our example there are 50 working days between Jan 15th and March 25th (no holidays during that time)
- This total number of days is divided equally into the number of distribution periods, four in this case two each with 10% and two each with 40%. In our example 50 days is divided evenly into four periods of 12.5 days each
- The total effort is divided based on each distribution profile %. In our example the first period (12.5 days) has 10% of the total or 100 hours, the second period (day 12.5 to 25) has 40% or 400 hours, the third period also has 400 hours and the final period has 100 hours
- Each period is mapped to a calendar month. If it falls entirely inside one calendar month the distributed hours are added to the total for that month. If the period in question spans more than one month then the effort is divided up evenly between the two or more months based on the number of working days from that period in each calendar month. In our example:
- The first period runs from Jan 15th to half way through Jan 31st so that's 100 hours in January
- The second period runs from half way through Jan 31st until Feb 18th so one 25th (0.5 / 12.5 of 400 hrs =16 hrs) falls into January while the remainder (400 - 16 = 384 hours) falls into February
- The third period runs from Feb 19th until half way through March 7th so that's 8 days in February and 4.5 days in March. 8/12.5 of 400 hrs = 256 hrs are added to February and the remainder (400 - 256 = 144 hrs) falls into March
- The last period runs from March 7th until March 25th so the final 100 hrs are entirely allocated to March
- The days allocated to each calendar month are added up and costed based on the labor rate applicable to your company's fiscal year for the calendar year and month in question. In our example the total hours for January would be 116 hrs, in Febrary 384 + 256 = 640 hours and in March 244 hours. Despite the bell distribution curve being "even" the fact that the task starts in mid-January and ends almost at the end of March explains why March's calendar month allocation of effort is more than double that of January.