Page 204 -

176 • Green Project Management

have a snack, or to tie their shoes), they catch up because their average pace
is faster than that of the kid in front of them. However, if a Scout who is
slower than the hiker in front of him pauses, he never regains his original
spacing and the line of hikers continues to expand. Herbie is slowing down
the whole group because Rogo (as the responsible adult) has to keep all of the
kids in viewing distance and has to call ahead for them to wait. Rogo gets the
inspired idea to place Herbie at the front of the line, which solves the prob-
lem of keeping the group together, since everyone has to walk as slowly as
Herbie does. But he doesn’t stop there. To increase the group’s rate of speed,
he redistributes what Herbie is carrying in his pack (cans of soda, a collaps-
ible steel shovel, and a jar of pickles) into the quickest hikers’ backpacks (as
well as his own). The lighter load allows Herbie to walk more quickly. Rogo
realizes that really, this is a system of hikers, hiking the trail together, and
that any individual hiker’s speed is dependent on the hiker’s speed in front of
him. He further realizes that the changes in pace are statistical fluctuations,
and Herbie is the system constraint. He has slowed down his speediest hik-
ers, and this intuitively seems quite wrong. However, the system is moving at
its fastest collective rate because it is moving at the constraint’s fastest speed.
So, even though hikers are not individually efficient, the system’s constraint is
2
efficient and therefore the system is running at its highest possible efficiency.
In critical chain project management, we do not have individual safety
buffers at each individual task; we instead move the buffering to the over-
all project date, to protect that team goal. Progress is measured by the use
(or lack thereof) of buffers.
Let’s translate this into action.
Here are the “five focusing steps” of the theory of constraints:

1. Identify the system’s constraints. What’s slowing things down?
Where is the bottleneck? Here are some triggers that help us identify
a constrained resource:
a. The resource is overloaded.
b. Work piles up in front of the resource.
c. Resources downstream from the resource under investigation
are idle some of the time.
2. Decide how to exploit the system’s constraints. Make sure there is
always work for the constraint to do. Don’t let it fall idle because of
lack of resources. This can be accomplished with a (small) “feeding”
buffer of work for the constraint. The responsibility of the resources
in front of the constraint is to make sure this buffer is always filled
with “just enough” work. Make sure that the constraint works only