Preferred method of Linking Task

[Kevin]

My VP of Operations and I are having a difference of opinion on how you link
task. He sets up schedules with SS predecessor links. I set them up with FS
redecessor links so that if a predecessor's duration extends then the
successor is push out accordingly. If you want the successor to start before
the predecessor is finished you use a negative lag.
I am trying to establish a company policy based on best practices for
building and updating MSProject for the construction industry and I need some
help. The items I want to establish policy around are: the way we link task
when building a schedule, how schedules are updated, do we or don't we use
the Actual Start and Actual Finish function when updating our schedules, If
not what is the preffered method.
Obviously everyone has there own way of building and managing schedules what
I am looking for is the recommended way.

 

[Jack Dahlgren]

In my opinion, both approaches are valid and you can and should use both of
them - as well as other types as well. There are subtle differences between
the way things are linked and the links should match as closely as possible
the way the work will actually occur. I tend to use a high percentage of FS
links. I also discourage use of lag (it is often hidden from view and has no
documention of what it is based on - eg: why -3 days instead of -5?). But the
"Best" way is not to just stick with one way. Be flexible and make your model
match how work is done using whatever modeling capabilities you have.

A policy stating that you must only use FS dependencies would be silly. Your
policy could describe the appropriate use of the different types of
dependencies with examples. That would be smart.

-Jack Dahlgren

 

[Kevin]

Jack thanks for the feed back. I agree the linking method need to match
closely to the actual work. What about using Actual Start / Actual Finish to
track and update the schedule. My experience is when you enter Actuals it
wants to break your links. I prefer to adjust my duration if I am ahead or
behind on individual tasked then show the % complete that way I see the
impact to my critical path. Do you agree or disagree? I know Actuals are part
of the system for a reason I am just not convinced that it lets me project
far enough in advance of activities to see the impact to milestones in time.

 

[John]

Jack thanks for the feed back. I agree the linking method need to match
closely to the actual work. What about using Actual Start / Actual Finish to
track and update the schedule. My experience is when you enter Actuals it
wants to break your links. I prefer to adjust my duration if I am ahead or
behind on individual tasked then show the % complete that way I see the
impact to my critical path. Do you agree or disagree? I know Actuals are part
of the system for a reason I am just not convinced that it lets me project
far enough in advance of activities to see the impact to milestones in time.

Kevin,
I just read your first post and Jack's reply. I agree 100% with Jack on
the linking. I noted he didn't address your question on entering actual
dates so I thought I'd throw my two cents in on that. Jack may have his
own or additional comments.

Remember that links between tasks are valid only until the successor
task actually starts. When you enter an actual start date, assuming it's
different than the planned schedule start, the link from the predecessor
task is null and void. Project doesn't break the link, the link is
simply ignored because it no longer has meaning.

Entering a date into the Actual Start solidifies the original plan by
turning it into reality. If everything goes per plan, entering a
non-zero value into the % Complete field will cause Project to
automatically set the actual start to the scheduled start. On the other
hand, if the task did not start as planned, entering an actual start
date causes Project to update the planned date to the actual date. If
you are thinking, "gee, that doesn't preserve my original plan". That's
right, and that's the reason for the baseline fields.

A similar thing applies to actual finish. Entering a date turns the plan
date into a real date. As actual dates are entered, Project
automatically adjusts the schedule, assuming it is not rife with
constraints, to reflect reality.

Hope this helps.
John
Project MVP

 

[Kevin]

Thanks John your reply was very helpful I am going to play with the actuals
for a while to see what your talking about.

 

[John]

Thanks John your reply was very helpful I am going to play with the actuals
for a while to see what your talking about.

Kevin,
You're welcome. Just a comment about semantics. In my world the term
"actuals" refers to actual cost data from our financial system. The
"actuals" term is not used when talking about dates.

If you have further question, post back.

John
Project MVP

 

[Steve House]

Just a note adding to the other comments. Remember that tasks are always
physical actions extending over time. Links describe the physical
relationship between the tasks so any of the four links types - FS, SS. FF,
SF - might apply depending on the physical nature of the two tasks. They
are not there to make the schedule look like something that someone has a
priori decided it 'ought' to look like nor are they subject to decision by
policy. Likewise, when actual progress is posted, it is supposed to
represent the date and time on which an actual, observable, physical event
took place. An actual start of 8/6/08 on a painting task means that if you
had been there, that is the moment you would have seen the painter pick up a
brush and start to apply the paint. As a historical physical event, it's
engraved in graniote once it occurs.

 

[Crit Chain]

Kevin,

Lots of good suggestions and comments, I would add to them "Plan the work -
Work the plan". Use as many FS as possible over the other options but, use
the FF, SS or SF when appropiate and avoid links on summary tasks as much as
possible.
Updating the plan during execution using the baseline and inputing actual
starts and actual finishes are key to tracking the plan to actual progress.
The real value returned is when you do the same job the second or third time
and can gain efficencies from reviewing the as accomplished schedule for
places we could have improved from last time.

Just a note adding to the other comments. Remember that tasks are always
physical actions extending over time. Links describe the physical
relationship between the tasks so any of the four links types - FS, SS. FF,
SF - might apply depending on the physical nature of the two tasks. They
are not there to make the schedule look like something that someone has a
priori decided it 'ought' to look like nor are they subject to decision by
policy. Likewise, when actual progress is posted, it is supposed to
represent the date and time on which an actual, observable, physical event
took place. An actual start of 8/6/08 on a painting task means that if you
had been there, that is the moment you would have seen the painter pick up a
brush and start to apply the paint. As a historical physical event, it's
engraved in graniote once it occurs.
--
Steve House [Project MVP]
MS Project Trainer & Consultant
Visit http://project.mvps.org/faqs.htm for the FAQs

My VP of Operations and I are having a difference of opinion on how you
link
task. He sets up schedules with SS predecessor links. I set them up with
FS
redecessor links so that if a predecessor's duration extends then the
successor is push out accordingly. If you want the successor to start
before
the predecessor is finished you use a negative lag.
I am trying to establish a company policy based on best practices for
building and updating MSProject for the construction industry and I need
some
help. The items I want to establish policy around are: the way we link
task
when building a schedule, how schedules are updated, do we or don't we use
the Actual Start and Actual Finish function when updating our schedules,
If
not what is the preffered method.
Obviously everyone has there own way of building and managing schedules
what
I am looking for is the recommended way.

 

[John]

Just a note adding to the other comments. Remember that tasks are always
physical actions extending over time. Links describe the physical
relationship between the tasks so any of the four links types - FS, SS. FF,
SF - might apply depending on the physical nature of the two tasks. They
are not there to make the schedule look like something that someone has a
priori decided it 'ought' to look like nor are they subject to decision by
policy. Likewise, when actual progress is posted, it is supposed to
represent the date and time on which an actual, observable, physical event
took place. An actual start of 8/6/08 on a painting task means that if you
had been there, that is the moment you would have seen the painter pick up a
brush and start to apply the paint. As a historical physical event, it's
engraved in graniote once it occurs.

Steve,
Graniote, is that a new type of substance that is in fact changed when
wormholes or other alterations to the space-time continuum are employed?

And you wondered if anyone really reads your posts

John

 

[Steve House]

Graniote = the smallest unit into which a block of granite can be subdivided
while still retaining the essence of its graniosity. Used as a substrate
for exotic engravings as it exhibits the permanence of the parent rock in a
much more convenient form-factor.

<grin>

 

[Mike Glen]

Nah - a Graniote is the coyote's cub's grandmother.

Hope that helps

Mike Glen
Project MVP

 

[JulieS]

<smile>

Great to have you back Mike!

Julie
Project MVP

 

[Mike Glen]

Thanks, Julie - we've just spent a week away in Pembrokeshire (West Wales),
on a cliff top facing the Atlantic! We suffered much rain and many gales -
but we're alright now


Mike Glen
Project MVP