5

I once managed a very intensive, very complex project that with every tool, illustrated an on-time schedule. But in the end, the project was severely delayed.

I understand the concept of "known unknowns", "unknown unknowns" and scope creep; but what is the community's guidance on the best way of anticipating the potential of schedule delays?

For instance, I understand the NASA Apollo project was subject to all the above (and more). How did the PMs anticipate delays that could have put them in Jeopardy of missing the Dec 1969 deadline?

Improve this question
2
  • 1
    Hi Fergus, I guess that I don't understand your question. what do you mean by the best way of understanding the potential of schedule delays.
    – Geo
    Feb 16, 2011 at 2:16
  • Thanks, I updated the question... it was somewhat ambiguous.
    – Fergus
    Feb 16, 2011 at 2:24

5 Answers 5

Reset to default
8

I think the best way to anticipate schedule delays is to do a thorough Risk Analysis and keep a Risk Register. Someone wise (@ashes999) once said, "projects don't suddenly fall behind schedule; they fall behind one hour at a time."

Keeping a Risk Register that identifies items which may affect the project schedule should prompt you to build in buffers when necessary.

Improve this answer
4
  • 3
    @ashes999's quote is awesome. That is a great, applicable quote to the world of project management, and with 20/20 hindsight I can see how true it is. +1
    – jmort253
    Feb 16, 2011 at 3:49
  • Risk management has almost nothing to do with time management. Risk register produces new activities that are managed in Schedule, but that's it. When Schedule Variance is there it's too late to think about risk analysis. It's time to think about "schedule reconciliation".
    – yegor256
    Feb 16, 2011 at 8:40
  • 1
    @yegor, it's true, according to PMI risk mgmt isn't related to time mgmt. @Fergus asked how to anticipate delays. EVM and sched variance are part of controlling the project...wrong application of PMI when asking about anticipating. We can use critical path/chain, what-if analysis (I like the Monte Carlo) as part of anticipating before we find out we're behind. I often like to think outside the PMI box and use other tools like a Risk Analysis to help predict the future.
    – CraigV
    Feb 16, 2011 at 14:16
  • Risk management has everything to do with time. A risk register should include mitigation plans (what to do when things go wrong) -- this effort requires additional time and effort. And always, always have time buffers for risks. That's how you avoid missing deadlines when risks actualize...
    – ashes999
    Feb 16, 2011 at 20:50
4

Let me tell you something that has been very useful for us.

We measure everything that we think might/could impact us with a Risk Analysis. This analysis tell us the things and the costs if they do impact, plus the probability. This takes care of the "known unknowns".

Then we take the total amount of the cost of the project, estimated during planning. And we take x% (depending on your budget and revenue streams), and we assign that for "unknown unknowns" Usually only sponsors will be able to use this money.

For the scope creeps, we use proper PM process to avoid them :)

Is not perfect but it works most of the time.

Improve this answer
2

There is a great indicator in Earned Value Management (EVM) called Schedule Variance, which is intended to spot schedule slippage.

Improve this answer
2

To accurately project how long a project will take, you need to understand that it is impossible to accurately project how long a project will take.

Despite years of experience, the most expensive project management software and quantitative techniques, projects go over schedule time and time again - keep this fact in mind and you can become pretty competent at estimating project schedules.

Improve this answer
2

You can't anticipate them. You can only make sure there's enough room in the schedule to accommodate them.

Keep scope well within 50% variance of the estimated due date. That is, if the due date for the project is 10 days from today, make sure the duration of the project is not more than 5 days.

Run Monte Carlo experiments on the schedule to best understand your exact time parameters and constraints.

For Apollo, they had the advantage that they could crash the schedule to the greatest extent possible. (My father-in-law worked on the Apollo project.)

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.