Hello J.Allman,
My comments are inline in your message below.
J.Allman said:
Julie, Thanks for the feedback but your answer ahs led me to
another
question. I was told by someone else that there are 4 rules of
updating
schedules:
"The four shalls:
There shall be no task with a start date left of status date with
0%
Complete, establish a new start date if necessary.
There shall be no task with a finish date left of the status date
that is
not 100% complete, establish a new finish date if necessary.
There shall be no task with a %Complete>0 with a start date to the
right of
the status date, you did not do the work in the future.
There shall be no task claiming 100% Complete with a finish date
to the
right of the status date, the latest this task can finish is the
status date.
If you do those four things, you probably have 80% of it covered."
____________________________
[Julie] Yes, I think Jim Aksel has this on his blog.
This makes me even more confused. I have a schedule that has a
primary
purpose of tracking work performed to detect any deltas such as
schedule
variance.
Something that is common and in conflict with what was written is
that,
unfortunately, it is not uncommon for tasks start dates to have
already
occured even though no work as started against the task.
[Julie] Not really. It is not uncommon for tasks to have been
scheduled to start in the past, but due to any number of issues, no
work has been done on the task. It is, however incorrect to say the
task started (work commenced) when it didn't.
Ex.
Test manager has a Task A with a start date of 6/6/09 and the
current date
is 6/10/09 and no work has been done on this task yet. This task
is
obviously running behind but there are other tasks behind it that
are
successors. If I was to follow the "4 Rules" then I would move
Task A to the
current date and that would inturn move the successors to the
right which
would in turn run up against the delivery milestone which has a
hardcoded
deadline.
[Julie] Yes, that is 100% true. But the alternative is to ignore
the fact that the task is now 4 days late and that if you do nothing
you most certainly *will* miss the delivery milestone. The purpose
of tracking is to enable us to see what is going on with the
progress in the project. If you are 4 days late, then as someone
who clearly doesn't wish to miss the milestone, you need to be aware
of the situation and take appropriate action. Not updating the task
as 4 days late is not going to fix the problem that you're late.
Are you purposing that I continually change all unstarted tasks to
reflect a
<start> date of the current day?
[Julie] I am proposing that you update the project including
modifiying actual and remaining work or actual and remaining
duration. Set your status date (Project > Project Information)
appropriately and after updating choose Tools > Tracking > Update
Project -- Reschedule uncompleted work to start after......
In my mind I make a formula for a RAG light that shows task A has
a variance
and have the Test manager address Task A to get it caught up (i.e.
Overtime,
Additional resource addded to Task A until it is caught up, etc.)
[Julie] Yes. But if the task was scheduled to start in the past, no
matter how talented your Test Manager is, s/he is not going to
change the fact that the task was not started. Update the task,
alert the Test Manager and now the Test Manager needs to come up
with ways (overtime etc) to get the finish date of the task to
adjust to meet the deadlines. Task A will still show a start
variance, but the idea is to minimize the finish variance and get
the schedule back on track.
If I am suppose to move unstarted tasks with start days to the
left of the
current day then please explain why my system is flawed.
[Julie] I hope I have
Thanks for your help and anyone elses input
[Julie] You're most welcome.
