Hello TOC

Fundamentals of TOC · Lesson 4 of 6

Constraints and policies

A constraint is whatever limits the system from getting more of its goal. Machines and markets can be constraints, but most often it’s a rule we set ourselves.

12 min · Not completed yet

What is a constraint?

A constraint is anything that limits a system from achieving more of its goal. It’s the weakest link from the last lesson.

Every system has at least one. If yours had none, it would make infinite money. And it has very few, because links rarely tie for weakest. Find them and you know where your leverage is.

Types of constraints

Constraints can sit in very different places:

  • Market: not enough customers want what you sell, at your price.
  • Resource or capacity: a machine, a team or a skill that can’t keep up with demand.
  • Material: not enough of an input to make what you could sell.
  • Supplier or vendor: a supplier who can’t deliver enough, or on time.
  • Knowledge or competence: too few people who know how to do a key job.
  • Financial: not enough cash to buy what you need to fill the orders you have.

And then there is one more type, which is the most common of all.

Policy constraints

A policy is a rule or a habit about how things are done: written down in a manual or simply “the way we work”. A policy becomes a constraint when it stops the system from getting more of its goal.

Goldratt found these everywhere. In his words:

“In most companies that I have analyzed, I have not found physical constraints. I have found policy constraints.” — Dr. Eliyahu Goldratt

He made the same point about markets, capacity and suppliers. Real market constraints are rare, but marketing policies that limit sales are common. True bottleneck machines are rare, but production and logistics policies that waste capacity are common. Supplier constraints are rare, but purchasing policies that cause shortages are common.

Some policies you will probably recognise:

  • Keep everyone busy. Releasing work just to keep people occupied floods the floor with work in progress and hides the real constraint.
  • Economic order quantity. Big batches lower the unit cost on paper but tie up cash and stretch lead times.
  • No overtime, ever. Even an hour on the constraint that would earn far more than it costs.
  • Cut costs everywhere. Including at the constraint, where an hour lost is an hour lost for the whole system.
  • Month-end push. Measuring sales by the month creates a rush at the end and idle time at the start.

Why bad policies survive

When you dig into where a damaging policy came from, you usually find that it made perfect sense when it was introduced. Maybe machines were expensive and labour was cheap, or orders were large and predictable. The conditions changed long ago, but the rule stayed. Nobody is wrong for having followed it; the job is to notice that its reason is gone.

Exercise: physical or policy?

  • The paint shop runs 24 hours a day and still can’t meet confirmed orders
  • Every order is released to the shop floor the day it arrives, so nobody runs out of work
  • Salespeople are paid on monthly targets, so most orders come in the last week of the month
  • Only one engineer is certified to sign off on the final test
  • Purchasing must buy in lots of 10,000 to get the best price
  • The plant has spare capacity and stock, but there aren’t enough customers who want the product
  • Any order above $5,000 needs three signatures before work can start

Your turn. Think of your own workplace. Write down one rule or habit that might be limiting results. Why was it introduced, and is that reason still true?

Check your understanding

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

1. A constraint is…

  • Any problem in the organisation
  • The most expensive resource
  • Anything that limits the system from getting more of its goal
  • A rule written in the policy manual

It’s defined by its effect on the goal. Many problems are real but are not constraints.

2. According to Goldratt, the most common type of constraint is…

  • A policy
  • A machine
  • The market
  • A supplier

Rules and habits limit far more systems than physical limits do.

3. Why do damaging policies survive for years?

  • Because managers ignore results
  • Because they never made sense
  • Because changing rules is illegal
  • Because they made sense when introduced, and the reasons were forgotten when conditions changed

That’s why the question “why do we do it this way?” is so powerful.

4. A company has idle machines, full warehouses and too few orders. Its constraint is most likely…

  • Capacity
  • The market, or a marketing policy
  • Material
  • Cash

When you can make more than you sell, the limit is in the market, and often in how the company sells to it.

5. How many constraints does a typical system have at one time?

  • None, if it’s well run
  • Dozens
  • Very few, often just one
  • One per department

At least one, or it would make infinite money, and very few, because links rarely tie for weakest.