Resource Allocation

[Martin]

There have been some interesting discussions on the
apparent conflicts between work and duration in a task. I
have a variation on the theme I would appreciate some help
with. We can get a situation where someone gets "behind"
with work and, in consiquence, Project extends the
duration of the task to make up. Is there any way in
which we can default to keeping the work and duration
fixed and increasing effort instead to enable them to
catch up? Or do I have to resort to a manual fix?

 

[Majid]

Hi, I had the same question and I am trying to see the FAQ
item # 37 but I am not able. Please advise

 

[James G]

Martin,

Your requirements can be fulfilled using a task-type of Fixed Duration.
Project will keep the original end-date and the work, (it just increases the
required units). This works fine, provided you do not "Reschedule Uncompleted
Work". However, you do have to be relatively careful in monitoring this
because it could easily catch-you-out...ending up with a masive resource
requirement all on one day!

HTH.

James.G

 

[Steve House [MVP]]

Consider though, how could you increase the effort if the resource was
already working at 100%? 100% means he will achieve 8 man-hours work of
work output during an 8 hour duration time period. It is physically
impossible for him to do more - there is no way to do 10 hours of work in an
8 hour time period, for example - thus he can't simply increase effort
arbitrarily unless he was assigned below maximum capacity to start with.

There actually are never any conflicts between work and duration but there
is occasional misunderstandings on the part of Project users on what the two
variables are actually measuring. "Work" is a measure of the total energy
expended and thus the output of the task. "Duration" is the time period
over which that energy gets expended. If we need to produce 100 widgets,
that's defines the work required. Working full speed we can produce 20
widgets a day - that output is what 100% resource assignment implies - and
so we estimate that it will take us 5 days to do it. So now we get to day
five and find that we only have 60 widgets on hand, we've only been able to
do 12 a day instead of the 20 we counted on. Is there any choice except to
extend the duration, work some more days, in order to produce those required
40 additional widgets? The task isn't done until we finish all 100, we
can't just abort at 60 and still complete the project. We can't go back and
retroactively increase the daily output - what's done is done. Up until now
we've been producing 12 a day even though we've been trying as hard as we
can - is it reasonable to believe the *next* few days we can crank it up to
20 per day or is it more likely that we'll continue on as we have been,
producing 12 a day? There was some reason that output has been limited to
12 per day and it's really not likely that will change overnight unless we
can find more resources to apply. I suggest that the most realistic
approach is to extend the duration by about 3.33 days.

Work, Duration, Effort -- why does changing one of them always change one
other? W=D*E is a fundamental project math identity that can never be
violated. It is the classic linear equation y=mx+b where two variables, an
independent variable x and a dependent variable y are related by a constant.
Set the independent variable and the value of the dependent variable is
determined by the equation. With Project you get to choose which of the
three terms - W, D, or E - is the constant by setting the task type.
Whichever remaining term you pick to adjust becomes the independent variable
and the laws of mathematics dictate that the third term MUST adjust so the
equation remains satisfied.