Unanticipated Events Screw Up Schedules


So after I posted the Probabilistic Scheduling post, I was working merrily away. I had made some small progress on the book, but was still finishing up other things. Finally, Wednesday I had cleared the entire day to work on the book. I was having trouble with one chapter, so I decided to make tea and do some timed writing. But I encountered an unanticipated event.

While picking up my electric teakettle, I fell down. Have no idea why, just fell. Normally, an ankle or knee gives out, and I fall sideways. I know to relax and go with the flow so I don't damage joints worse than they are. But this time I fell straight down. Did a Greg Louganis on the table in my office.

Head wounds hurt. Almost as much as childbirth. And when they bleed, they bleed a lot. Called the doctor, breathing through the pain, was told, “Go to the ER.” Ok, got in the car, drove myself to the emergency room, waited several hours for my 4 stitches, and returned home.

I knew I was not myself; I didn't even read in the ER. (I read all the time. Everywhere. Unless I'm with other people. But breakfast doesn't count.) But by the time I returned home, I was in much better shape than when I left. The local painkiller was still working. I was no longer bleeding. And my headache was gone because of the local. I was still shook up, but certainly ok.

So, Wednesday was a fairly lost cause, at least for writing. I made more progress yesterday, and hope to do more writing this afternoon, after I take care of other little things. And yes, I'm fine. I have blue stitches on the back of my head. Mark tells me no one can see them because of my hair. Phew!

Unanticipated events happen all the time. The bigger and less cross-functional your project, the less impact they have. But that's because you're barely making progress as is. The smaller the project, the more cross-functional the team, and the more people depend on each other, the more impact these events have.

It's not possible to plan for these events, except to have some contingency or slack in the schedule.

For several years, when I was scheduling projects, I would assume we would lose time to weather events in the winter and power outages in the summer. I didn't know when they would happen. I was sure they would. Those are anticipated events, even if they're unscheduled.

But unanticipated events can be happy, such as a key person getting married on the spur of the moment. (We had a quick party, he left and returned a week later after a honeymoon.) Or, they can be injuries, such as mine. To be honest, an afternoon of lost work is not the big a deal in the scheme of things.

So when you schedule, remember those unanticipated events. Give yourself enough of a contingency plan to deal with them. If you're working in an agile lifecycle, your velocity will self-correct. But don't assume they're not going to happen. Something will. You just can't tell what.

4 Replies to “Unanticipated Events Screw Up Schedules”

  1. Queuing theory suggests that you should plan work at about 50-60% of work capacity to avoid bottlenecks. The theory says that you will get more work done this way than if you plan at 70-80% of capacity. I think some kind of fudge-factor to account for the historic “unaticipated problem rate” for your projects is not a bad way to handle this. If you put this extra time as a buffer at the end of your project, then manage against the remaining buffer size, you can leave yourself the opportunity to finish ahead of schedule if you don’t need it.
    Get well soon.

  2. I think one thing can be done. Learning from past. Looking on the past projects you can at least estimate some kind of average time spent on unexpected issues.
    However, more often I see schedules screwed up by not planning obvious events. If you’re interested in reading a bit more here’s the link.

  3. We do an exercise with groups to see just how much time they really have available for their projects – not counting contingencies. Most project methods suggest allocating at 80% and it is useful to start with a more realistic baseline. We start with the maximum number of hours in a week and then subtract time for the usual – a weekly allocation for vacations, sick time, holidays, administrative time, phone calls/emailing, fire-fighting, and, oh yes, those things called ‘life’ (eating, sleeping, family time). The median available time left for project work?
    16 hours – and that varies from 25-30 hours for individual contributors working on just one project to negative time for project managers juggling multiple efforts.
    Yes, plan on the unexpected – AND, most usefully, expect the expected.
    Hope you heal quickly.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.