I do the same situation every month. It is tedious.
We maintain a single master.mpp with external predecessors. Each month my
subcontractor submits their monthly schedule (without the links to the
master). That is, my sub has no clue that I integrate into a master file and
use their file to drive my dates. The key is to keep all the file names the
same. Master is always Master.mpp and my sub is always SubProject.mpp. So
when my sub submits a new file I use the file compare utility (gulp -- it
does have its quirks). Once I am satisfied that the new file submitted is an
accurate reflection of their progress, I overwrite the old SubProject.mpp
file with the new one. Since the file names stay the same, when I open
Master.mpp it is happy to look for the external links in the same location.
Again, my SubProject.mpp has no external sucessors and is not driven by
Master.mpp. It only flows one way. Of course we keep appropriate monthly
archives.
It appears you may be able to do something similar.
Another approach is to have a section at the top of each file called Givers
and Receivers. Givers are the ONLY exit from the file. A giver must have an
internal file predecessor and an external successor to a differnt file.
Receivers are the opposite. A Receiver must have external predecessors and
must drive one or more internal successors. The ONLY place an external
predecessor may enter the file is at the point of a receiver.
But you may get more complicated than that because of your file depth. In
cases like yours, I establish a unique number scheme in every file and store
it in a text field. To keep it simple let's just call every file with a
letter name: A, B. C... I then create a formula in every file in a text
field. The formula for file "A.mpp" is: "A"+[Unique ID]. For file "G.mpp"
the formula would be "G"+[Unique ID]
Then, I use either another text field or the Notes area for each giver and
receiver. If I am in File "B" and I have a receiver from file "D" then I put
the link number in to the notes. In file "B" as a note I may say "D123"
which means the true predecessor is in FIle D.mpp and is UNIQUE ID 123. Why
use Unique ID? You will see the Unique IDs actually shift when you move to a
master file, so I invented the Text field formula technique, the formula
value remains constant. Also, Task IDs can change but a Unique ID never
will. Suppose a sub adds tasks above your link?
What this does is give you a method to find the real cross references on a
manual basis. Trust me, you will need it.
Implementation: Mandate that all suppliers may not delete these type of
tasks (if I delete a task and re-enter it, the Unique ID changes ... oops!).
You have to get them to commit to the formula scenario.
Needless to say, it requires some thought. We maintain a spreadsheet of
Giver/Receivers under configuration control. No one is allowed to change the
date of a giver or receiver without going to the Configuration Control Board
for permission. This is how we coordinate impacts since file owners can be
separated by continents (like England and the USA). If you need some
additional help, you can eventually dig out my contact information from the
blog if you poke around.
Not to give away my trade secrets, but you can make up your own numbering
scheme on the A123 type things. You may want to tie it to a WBS, a phase of
the contract, etc. Configuation control is your key. You are verging on
using Project Server to control all this along with some access rights. The
problem is going to be an IT issue with getting your subs through your
firewall to update their files. We have found CITRIX is a useful tool for
this scenario as well.
--
If this post was helpful, please consider rating it.
Jim
Check out my new blog for more information:
http://www.msprojectblog.com
DavidC said:
Hi Jim,
Excuse my breaking in to your response, but this is an issue I have been
struggling with over the past 18 months.
Background:
The project is a construction project which starts with our schedule
predicting the full project timeframe. Then when the contract is let we have
effecively two other schedules for the construction, and these are linked
together and into my overall schedule, later another three contractors are
engaged by the main contractor, and their schedules are then integrated into
the contractor's overall schedule. At this stage we now have my schedule,
with the contractor's schedule integrated into mine, and the contractor's
schedule having four schedules from their contractors integrated in. I then
have a further two contractor schedules which need to be integrated into mine
and some links between the other contractor schedules. This rather
convaluted contract arrangement is simply the nature of the construction work
and the split in the scope that has been made. It does actually make sense.
The issue is, as you alluded to, that these master/submaster projects are
thwart with dangers from spurious links being created when individuasl
schedules are moved. Problem is that each contractor needs their own
schedule to update and manage and not necessarily see the other schedules.
Often issues of IP and confidentiality impact on what can be viewed.
I am therefore trying to find a solution which enables me to have a full
schedule for the project that each contractor can manage independantly but
does not leave old external link refereneces. My thought is simply to have
the updates completed on a central server that everyone can access and limit
the read options, and then each contractor can save their own copy outside
the server in their own system if they wish. They then amy have issues with
external links left in the schedule impacting on how their schedule then
reflects progress.
I am sorry to impose like this but often two or more heads are better than
one, and it is easy to get fixed on one course of action.
Many thanks
DavidC
