Statistical Fluctuations Can Disrupt Production

This article is an excerpt from the Shortform book guide to "The Goal: A Process of Ongoing Improvement" by Eliyahu M. Goldratt. Shortform has the world's best summaries and analyses of books you should be reading.

Like this article? Sign up for a free trial here .

What are the statistical fluctuations in production? Is there anything you can do to reduce or prepare for them?

Statistical fluctuations are the natural deviations in production rates. There are always things that cannot be predicted and these factors impact production.

Keep reading for more about statistical fluctuations in manufacturing and what they mean for your production rates.

Statistical Fluctuations in Manufacturing

In manufacturing, a balanced plant tries to match average capacity of every resource exactly with market demand. Any resource beyond the average rate is seen as extraneous, so it is either put to use or eliminated. This is the traditional mode of thinking at the protagonist’s company in The Goal.

There are some things that can disrupt production. For example, statistical fluctuations because many factors cannot be predicted precisely.

Even at an average steady state rate, there are statistical fluctuations in production. Someone may produce 2 widgets per minute on average, but at times he produces 2.5 and at times he produces 1.

Larger events – machines may break down; workers many get sick; inclement weather may arrive.

Statistical fluctuations happen regularly at each part in the chain. However, each downstream part can only catch up to the extent that the upstream part permits it to. Negative fluctuations bring down every later step of the chain; positive fluctuations are constrained by the next bottleneck. Over time, this causes a lower than expected average throughput.

Analogy of the Hiking Line

Imagine a troop of 10 boys hiking single-file on a narrow trail in the woods. The leader of the pack sets a comfortable pace that everyone on average should be able to meet.

Every boy is only able to catch up to the boy in front – he can’t pass the boy in front. Thus, the speed of each boy is constrained by the boy in front.

Analogy to manufacturing: the first boy is the most upstream step; the last boy measures throughput; the distance in between is inventory.

Here’s an example of a negative fluctuation

Say the 3rd boy in line stops to tie his shoes, increasing the gap between him and the 2nd boy. The 4th boy runs into the 3rd boy and has to match his slower pace. Even if he could walk faster, the 4th boy is constrained.
The 3rd boy gets back up and hurries to make up the distance to the 2nd boy. But the 4th boy is already tired and can’t jog, thus increasing the lag.
These fluctuations happen at different points throughout the chain. The 6th boy gets tired and holds up the 7th boy, who is energetic but can’t get past the 6th. Then as the 6th boy regains steam, the 7th boy gets tire and lags behind.
Inevitably, even though all the boys are at the same average rate, the gap between first and last boy increases. The fluctuations are accumulating because the ability to go faster than average is restricted by your upstream step.
This gap can be closed only if the last boy – and all the boys in front of him – all move much faster than the first boy.

Let’s say to relieve these pesky dependencies, you order all the boys in speed, with fastest leading the pack. All boys can now move unconstrained and operate at their individual peak efficiencies. However, the distance between the first boy and the last boy will incrementally grow, unbounded.

This is what happens when you concentrate on single-step efficiency without focusing on throughput. All steps are working at full steam, but inventory progressively increases. The fastest step is churning out work ceaselessly, but the other steps can’t catch up.

To solve this accumulating gap, let’s try a new solution: order the boys with slowest first and holding up the line for all other boys. On the surface, this seems very inefficient – the other boys are nowhere near their peak efficiency, and they’re all running into the next boy.

However, there are a few key advantages:

The bottleneck is now quite obvious. We can now increase throughput by speeding up the slowest boy in front. For instance, we can redistribute the slowest boy’s backpack load to the other boys.
Inventory – the distance between the first and last boy – is now dramatically reduced. All boys are producing just enough to match the pace of the first slow boy.

This wonderful analogy for statistical fluctuations should make clear the flaws of defining the average of all steps, without regard to the overall throughput and the bottleneck steps.

Statistical Fluctuations Can Disrupt Production