Early Start trumped by Actual Start.

K

ken nj

I'm trying to come up with a way to determine which tasks were started
later than they should have been. Early Start is exaclty what I want
because it shows the earliest date that a task could start based on
the predecessors contraints. However, once an Actual Start has been
entered, Early Start takes on that value (can i block this behavior
somehow?). I want to create a report (I am programming asp pages to
run against the sql server db) and i would like to show which tasks
were started late compared to when they could've started. Also, I am
showing which tasks took longer than they should have, ie Actual
Duration more than Base Line Duration. Upper management here wants
to do project post mortems to determine where more resource need to be
focused.

-ken
 
D

Dale Howard [MVP]

Ken --

To accomplish what you wish to do, you need to save a baseline for your
project before you begin entering actuals. Click Tools - Tracking - Save
Baseline to accomplish this. To view your Start date variance for the
project after the project is underway, apply the Tracking Gantt view (View -
Tracking Gantt) and then apply the Variance table (View - Table - Variance).
Examine the Start Variance column for the information you wish to see, which
compares the current or actual Start date against the origninal planned
Start date, which is the Baseline Start date.

To assess Duration variance, there is no default Table that displays this
information, therefore, you would need to create your own Table to do this.
Follow these steps to create a Duration Variance table:

1. Open any project
2. Click View - Table - More Tables
3. Select the Variance table and click the Copy button
4. Name the new table something like "Duration Variance"
5. Select the "Show in menu" option
6. Delete all of the Start and Finish date fields, leaving only the ID and
Name fields
7. Insert the following fields:

-- Duration
-- Baseline Duration
-- Duration Variance
-- Actual Duration
-- Remaining Duration

8. Click OK when finished
9. Click the Close button (not the Apply button)

If you are NOT using Project Server, then after you have created this new
Table, you will need to copy it to your Global.mpt file using the following
steps:

1. Click Tools - Organizer
2. Click the Table tab
3. From the list of Tables on the right, select the Duration Variance table
and click Copy
4. Click the OK button

If you are using Project Server 2002 or 2003, use the Organizer to copy the
table from the current project to the Enterprise Global file by completing
the following steps:

1. Leave the project open that contains the Duration Variance table
2. Click Tools - Enterprise Options - Open Enterprise Global
3. Click Tools - Organizer
4. Click the Views Available In drop-down list in the bottom left of the
dialog and select the current project
5. Make sure the Views Available In drop-down list in the bottom right
shows the Enterprise Global
6. Click the Tables tab
7. From the list of Tables on the left, select the Duration Variance table
and click Copy
8. Click the OK button
9. Save and close the Enterprise Global
10. Close and relaunch Microsoft Project Professional

To apply this Table, apply any task View such as the Tracking Gantt view,
and then click View - Table - Duration Variance. Hope this helps.
 
K

ken nj

I appreciate the feedback but it isn't quite what I need. I
understand baselines, baseline tracking, variances and all that.
Here's is what I'm trying to do. We are using project to track
product launches. There will be hundreds to 1k per year projects that
are ultimately tracked. Each one has 30-40 tasks based on templates.
The are more than one critical path through the project. Upper
management want to be able to track not only which tasks were
completed 'late', (compare actual finish to baseline finish), but also
which tasks where the root cause in causing other tasks to be late.
This could be caused by two things, Tasks that took too long to
accomplish (actual duration greater than baseline duration, and also
Tasks that were started later than they should have (probably because
a department did not communicate well with another department or the
department did not have enought people to work on it). I refer to
this date as the 'should've started date'. I found the 'early start
date' for a task computes this date by analyzing the predecessors and
lag times. However, once an actual start date is entered, the early
start takes on that value. I could analyze the predecessors field and
compute the value my self but surely there's an easier way.

-ken
 
D

Dale Howard [MVP]

Ken --

You are asking Microsoft Project to do something is not programmed to do.
When an Actual Start date is entered for a task, then how could the Early
Start field possibly show anything else other than the Actual Start date?
 
K

ken nj

You are asking Microsoft Project to do something is not programmed to do.
When an Actual Start date is entered for a task, then how could the Early
Start field possibly show anything else other than the Actual Start date?
Click to expand...

here's the definition of 'Early Start Date' from the db documentation:

"The earliest date that a task could possibly begin, based on the
early start dates of predecessor and successor tasks, and other
constraints"


I like the first part of the definition but the end essentially means
whatever MS wants it to mean. What i need is what I call the
'Should've started date' or the date the task should have started
based on the predecessor. I want to compute the difference between
this date and the Actual Start date to show how much time was wasted
before a task was started. maybe it should be called the 'Task
Procrastination Duration'. Whats strange to me is that the 'Early
Start Time' captures this exaclty the way I want until an Actual Start
Time is entered.


-ken
 
D

Dale Howard [MVP]

Ken --

I appreciate what you are attempting to do. Unfortunately, I do not know of
any other way to accomplish what you want other than comparing the current
state against Baseline values. Perhaps someone else has an idea for you.
Sorry.
 
Top