Critical Task with Slack

[Bill Swihart]

I can not get a task that has obvious slack to show up as non critical.

I have a task (T3) that is constrained to start no sooner than... and is
linked to a critical task (CT4). The critical task is driven by another
task (T2) that sets the start date of CT4 well beyond the finish date of
T3. T3 shows as critical. If I remove the constraint of T3 to ASAP then it
becomes non-critical but does not start when I need it to.
Moving it back causes it to show critical. Suggestions please.

Bill

 

[JackD]

Does it have a deadline? Is there any lag on the dependency? What sort of
dependency are you using? Why are you using a No earlier than constraint?

 

[Bill Swihart]

The specific task does not have a deadline. the project does.

There is no lag in the dependency.

It is a finish to start.

Because it does not have a predecessor to drive it and this is when it is
scheduled to begin.

 

[Jan De Messemaeker]

Hi,

Insert a column "Total Slack".
Does it show zero?
If yes, the slack is zero and the task is critical, whatever the
appearance - Noone can say without seeing what may be the reason.
(Could it be you use Must Start on instead of No Earlier Than?)

If it isn't go to Tools, Options, Calculation, Tasks are critical when slack
is... should show a figure othe rthan zero; put to zero.
HTH

 

[Steve House [MS Project MVP]]

Why is it scheduled to begin at that SNE constraint date? Unless there is
some external factor, parts not available before a certain date for example,
one usually wants tasks to begin as soon as possible so that the project is
finished in the shortest practical time period, as far ahead of deadline as
practical so that the inevitable delays etc that creep in don't cause you to
blow by the deadline.

You wrote "...this is where it is scheduled to begin" suggesting that the
schedule is determined in some manner other than through the use of MS
Project's calculations. But IMO that seems to be putting the logic
backwards and contrary to the basic purpose of using scheduling software
such as Project in the first place. Its function is not to simply
illustrate a schedule that you have derived through some external process.
Rather, its job is to calculate for you the optimal schedule that will let
you finish your project on time and within budget given the work that has to
be done and the assets at your disposal to do it with. You don't input the
schedule into it, instead it outputs the schedule for you, just the opposite
of what it sounds like you've done.

Just my $0.02

 

[Bill Swihart]

Let's say that the resource to perform this task will be hired on that date
and then the task can begin. That is a legitimate contraint on when the
task can begin. there is plenty of time to perform the task and indeed it
is being performed ASAP prior to the next task which is critical. Your
suggestion would have me hire the resource too soon thus wasting $.

 

[Bill Swihart]

Good suggestion, This will allow additional troubleshooting.

 

[Bill Swihart]

Problem resolved and thanks for the suggestion to review total slack. The
problem was created because the deadline for the project could not be
supported thus creating the requirement that all tasks supporting the
deadline become critical.

Good Exchange! Happy news grouping!

 

[Steve House [MS Project MVP]]

That was the nub of my question and you are quite correct, that is a very
appropriate use of the SNET constraint. I asked because many people
introduce constraints in an attempt to get the schedule to conform to some
pre-conceived idea of what it "should" look like and almost invariably such
a strategy ultimate creates far more problems than it solves. I wonder if
this might be a situation where a SF link would work well. We have tasks in
a chain A->B->C->D and also task B1->C, B1 being that task the we're talking
about that the consultant is on. B1 needs to finish just in time for task C
to begin, C's timing being driven by the sequence A->B->. Would it work for
you to schedule C as the predecessor with B1 as its *successor* with a
Start-to-Finish link? I think of "predecessor" meaning the task controlling
the timing and "sucessor" meaning the task whose timing is controlled in
this sort of context, the idea being that the start of C determines when B1
must end and so in turn that determines the latest B1 can start.

--
Steve House [MVP]
MS Project Trainer/Consultant
Visit http://www.mvps.org/project/faqs.htm for the FAQs

 

[Bill Swihart]

This will also work, thanks for your help.

That was the nub of my question and you are quite correct, that is a very
appropriate use of the SNET constraint. I asked because many people
introduce constraints in an attempt to get the schedule to conform to some
pre-conceived idea of what it "should" look like and almost invariably such
a strategy ultimate creates far more problems than it solves. I wonder if
this might be a situation where a SF link would work well. We have tasks in
a chain A->B->C->D and also task B1->C, B1 being that task the we're talking
about that the consultant is on. B1 needs to finish just in time for task C
to begin, C's timing being driven by the sequence A->B->. Would it work for
you to schedule C as the predecessor with B1 as its *successor* with a
Start-to-Finish link? I think of "predecessor" meaning the task controlling
the timing and "sucessor" meaning the task whose timing is controlled in
this sort of context, the idea being that the start of C determines when B1
must end and so in turn that determines the latest B1 can start.