Schedule Quality - Essentials for reporting

[Bruce McF]

Are any code examples available for basic schedule quality? I am only trying
to quantify how well each schedule is maintained with "complete in the past",
"incomplete in the future" and "resources as densely scheduled as possible
witho utoverallocation". (A check for resource assignments on all detail
tasks is implicit in this as is last status update, etc.) I am trying to
generate metrics to tell us if we can trust the schedule information and to
set minimum-acceptable quality limits.

I am not trying to check for any best practices at this level ... only that
the Finish Dates are credible. If tasks are well scheduled relative to the
status date, I figure that the Finish Dates are credible and a reasonable
estimate of when tasks will complete (within typical uncertainties). That's
the essential bit I need for reporting. (Best practices tend to make it
easier to maintain the schedule, but I am interested in how useful the data
is after normal processing.) I call this FDI ... Finish Date Integrity.

I think I've got the code to tell whether statused tasks fall on the correct
side of the status date and am using TSV's to look at availability vs. plan
to spot over/under allocations ...

BUT I HAVE TO THINK THIS HAS ALREADY BEEN DONE A HUNDRED TIMES?

I suppose if I could turnoff all the best-practice checking in the
commercially available tools, that would work, but my first reaction is that
we are pretty close with our code to date and will have more direct control
of the metrics that would be useful in our applications.
(resource-constrained iterative software development).

I know you all are loaded and appreciate any guidance you might offer.

Thanks,
Bruce

 

[Rod Gill]

Whilst I have coded the occasional tool to help with this, the commercial
tools do cover everything you might need to know or never will! EXCEPT: Have
a copy of the key deliverables for the project been copied and pasted to the
schedule and all tasks required to complete them scheduled? Has the schedule
been peer reviewed for credibility? Is the Critical path accurate and "fit"
what people expect it to be?

Plus, each project and program of projects are different, so I suspect you
won't find a tool or existing macro that exactly suits your governance
requirements for your current projects and management needs!

--

Rod Gill
Microsoft MVP for Project

Author of the only book on Project VBA, see:
http://www.projectvbabook.com

 

[RobV]

Bruce, as you suggest, this has been done before--there are commercial
packages that provide the information you're looking for. I also agree with
Rod that you will probably not find a single tool that does all and only what
you need. Nonetheless, a quick Internet search would be worth your time.

I suggest that you look for tools with a sound theoretical basis--something
like Earned Value or Earned Schedule. They provide a quantitative assessment
of what you call FDI, even at the task level.

 

[Bruce McF]

Rod, Rob:

Thanks. I have (had already) downloaded one trial and will make another
pass on the internet. Thanks for the advice.

In our case, CP doesn't drive delivery date and the order of work packages
doesn't matter ...except for (relatively) short sequences. Our problem is
more like scheduling a series of different products to be produced in a
chemical plant (across a number of specialized chemical processing units) in
a year ... except that we wait to send the products (functionality) to market
in one big batch (release). Time-boxed and resource-constrained.

It is a matter of assessing Finish Dates relative to Target Dates and
understanging that variance in the context of the natural uncertainty in the
schedule. If we are well scheduled, the uncertainty is only the uncertainty
in the estimates themselves. If not, additional uncertainty is added in what
shows up in the reports based on poor scheduling.

Thanks! I will look more to the internet ... but continue with my TSV's in
parallel ... as suggested.

Thanks!