Hello TOC

Advanced Critical Chain · Lesson 2 of 8

The T20 simulation

Two tiny projects, one shared resource, and a stopwatch. Feel what multitasking does to lead time, quality and capacity, then make it harder and watch it get worse.

20 min · Not completed yet

Internal wastage is both bad news and good news. Bad, because it’s our own doing. Good, because what’s in our control can be fixed. So where does it come from in a multi-project environment?

Good and bad multitasking

Driving in heavy traffic while negotiating a deal on the phone is multitasking, and everyone knows it’s risky. But what about a shared resource expected to work on tasks from several projects?

Almost every project has some limited resource. If nothing else, every project shares the bandwidth of senior management. Yet this is rarely recognised, and every work front is opened as soon as possible. With too many projects running at once, resources are pushed to work on more than one task at a time. Bad multitasking becomes unavoidable.

Play: the T20 simulation

Before you play

The projects. Project P1: write the numbers 1 to 20. Project P2: write the letters A to T. Both are urgent, and you are the only resource for both.

Round 1, multitasking. Priorities keep shifting, so you switch after every step: 1, A, 2, B, 3, C… until 20 and T.

Round 2, one at a time. Finish P1 completely, then do P2.

How to play. Each project has its own keypad, with its symbols in shuffled positions. Tap the next symbol in order; what you’ve written shows above the keypad, like a page. The clock starts at your first tap. Go as fast as you can.

Then make it harder. Rounds 3 and 4 use odd numbers, every other letter, and a third project of repeating shapes, as in the classroom version.

What the results show

Plot both rounds on a time scale. Most people find:

  • The experience. The multitasking round felt chaotic, confusing and stressful.
  • Lead time roughly doubled. In the single round, P1 reached its customer in about half the time.
  • More mistakes. Overwriting, striking out, tapping the wrong key.
  • Lower output. In the time multitasking took, you could have finished at least one more project.

Why did lead time go up so much? Two reasons.

The larger one is non-productive time. While you work on P1, P2 sits in a no-work zone, and vice versa. You were working continuously; the projects were not. If projects could speak, they would say: someone worked on me, then nobody did, then someone did again. Which do we want: resources busy all the time, or projects being worked on all the time? The second.

The other is set-up and set-down time at every switch. The human brain can’t process two things at once. At every switch it puts one task down and picks the other up, recalling where it was. Remember reading a book over a week: even with a bookmark, you reread a paragraph or two to reconnect with the story.

Make it harder

Writing 1 to 20 is a no-brainer, so quality barely suffers. Real projects aren’t that simple. In rounds 3 and 4 the work needs more thought, and a third project joins in. Most people find the multitasking round becomes far more chaotic, with many more errors. Someone could protect quality by giving up on time, because it’s only a game. In real life, giving up on time isn’t an option.

The more complex the work, the more cognitive effort it takes, and the more errors multitasking causes. Errors mean rework. And rework acts like a catalyst in a chemical reaction: more rework, more switching, more rework, a vicious loop that never ends.

The hidden damage: work moves in waves

Imagine two departments: yours upstream, passing work to department 2 downstream. If you multitask, department 2 waits a long time for anything, and then both projects arrive at once. Its load swings between peaks and nothing. Work moves in waves, and the system loses capacity downstream too.

Pause and think: many organisations believe that starting a project as early as possible makes it finish early. What did the T20 simulation show?

  • Early start does lead to early finish
  • Start date doesn’t matter at all
  • Early start doesn’t lead to early finish; in a multi-project system, a later, focused start often finishes earlier
  • Projects should never start

In the single round, P2 started later than in the multitasking round, yet both projects were done sooner. Starting everything early just adds to the switching.

To sum up: bad multitasking is the root cause of most internal wastage in a project environment. It creates chaos and constant stress, stretches lead times, sends work in waves, triggers unnecessary expediting, wastes capacity, and breeds rework.

Check your understanding

Answer all five. Get four right to complete the lesson.

1. The largest source of extra lead time in the multitasking round is…

  • Slower writing
  • Non-productive time: each project waits while the other is worked on
  • The stopwatch
  • Errors

Resources are busy; projects are idle half the time.

2. Set-up and set-down time at each switch comes from…

  • Moving between desks
  • Software loading
  • Meetings
  • The brain recalling where it left off, as when rereading a page after a bookmark

The brain can’t process two things at once.

3. As work gets more complex, multitasking causes…

  • More errors, and so more rework
  • Fewer errors
  • No change
  • Faster completion

And rework feeds a vicious loop of more switching.

4. When an upstream department multitasks, the downstream department…

  • Gets a steady flow of work
  • Has nothing to do
  • Waits long, then gets several projects at once: work moves in waves
  • Becomes faster

Feast and famine downstream too.

5. Which should we aim for?

  • Resources busy all the time
  • Projects being worked on all the time
  • Both equally
  • Neither

Project flow, not resource busyness, decides lead time.