[Total: 0    Average: 0/5]

The concept of TOC was introduced under the analogy of the chain : “A chain is a sequence of strong and weak interdependent elements. Improving the performance of the chain can only do this by improving the weak element“.


It is a method developed in the 80’s by Eli GOLDRATT. Physicist and philosopher by training, this work focused on solving the problems of business organizations by applying the scientific method.

It will be through the development, in the 1970s, of a scheduling software based on the algorithm called OPT (« Optimized Production Timetables ») that GOLDRATT developed his philosophy.

It’s via his book « The Goal »1, that he made this method known. She was at the time much criticized 2 . Many people have said that this was just common sense, and that this theory did not help. They thought this theory was only about finding the Herbie, then the next, and so on. And especially, at the time, this methodology was against the MRP method, the most recognized and used method3 of the time.


The theory of constraints goes further. It highlights the way to build a system that improves and manages constraints. It is not a question of focusing on the levers of progress of the system but of understanding how these levers have an impact on the rest of the system. The theory of constraints4 :

  • challenges many principles about how we manage our businesses and organizations.
  • is based on the logic of cause and effect and was derived from the exact sciences. He calls this the Thinking Process.
  • Relies on Little Law to demonstrate that the control of the system is based on the bottleneck.

What is a constraint ?

A constraint is the process step that limits the production of the value chain and therefore defines its output speed.

There can only be one constraint per value chain.  

The notion of Bottleneck

By definition, a constraint is a step in a value chain that limits its throughput. It does not allow the system to perform better. APICS gives the following definition :

« Any element or factor that prevents a system from achieving a higher level of performance relative to its purpose ».

Starting from this definition, there can be an infinite list of constraints depending on the context. Here are some examples :

  • Material and human resources.
  • Knowledge or skills.
  • Methodological see even paradigms.
  • Customers or suppliers.
  • Time or location.

The motto of the Theory of Constraints


1 hour lost on the constraint, it’s 1 hour lost for the whole system.  

The challenge is to identify the constraint of the system to apply the leverage effect. By focusing on removing the constraint, we can pass a performance gap.

The challenge is to identify the constraint of the system to apply the leverage effect. By focusing on removing the constraint, we can pass a performance gap.

The intuitive reaction, the fastest first, is not the right answer. Herbie, the slowest scout (in the book of Eli GOLDRATT), when placed in the lead, ensures the fastest movement of the entire column. But back to the bridge :

  • A =1 mn
  • B=2 mn
  • C=5 mn
  • D=10 mn

The statement states that no more than 2 pedestrians can be on the bridge at the same time. Here is the solution: CDBA. C and D leave together. We count the time as soon as they leave. C (5 min) arrives at the end of the bridge at 5 min, D (10 min) is then somewhere on the bridge. B (2 min) then leaves and arrives at the end of the bridge at the 7 mn, D (10 min) is still somewhere on the bridge. A (1 min) then leaves and arrives at the end of the bridge at the 8 min. D (10 min) is for 8 minutes on the bridge and ends up arriving on the other side at the minute 10. Total time: 10 min.

The 9 rules of Theory of Constraints

Rule 1: Balancing flows and not capabilities

It is impossible to perfectly balance * the capabilities of a company. We must look for a 100% occupation of the resources but look for a fluid flow. We must therefore calibrate the flow sought on that of the neck and to meet the customer demand.

*Why it is impossible to perfectly balance flows :

  • Demand often varies.
  • Production is a sequence of random events.
  • The capabilities of equipment generally vary by discrete values.

Rule 2: Any Loss on Stress is a Loss Throughout the System

By definition, the bottleneck has no capacity reserve, it can not make up for lost time. Any loss of time on the neck is therefore a permanent loss.

Rule 3: An hour won on a non-trick is only a decoy

Resources remain dependent on the bottleneck. An hour gained on a non-bottleneck, does not benefit the system, but will increase stocks.

Rule 4: The level of use of a non-bottleneck is not determined by its potential but by the constraints of the system

Non-bottleneck resources must produce at the rate of the bottleneck. The use of non-bottlenecks according to their capacities is only overproduction. Thus, a bottleneck machine that feeds a non-neck machine must not be activated at 100% but only at what the neck can provide..

Rule 5: Use and Activation of a Resource Are Not Synonymous

A resource must be activated at 100% but used at the level of the need of the neck. Activation is defined as the use of resources without the need for Throughput. The use, it is defined as the use of a resource by necessity for the Throughput. Thus, a non-neck machine that feeds a neck machine must be “ready” to produce at any time depending on the neck. But must not produce more than what the bottleneck can do.

Rule 6: Bottlenecks Determine Both Release Rate and Inventory Levels

The overall rate of the system can not exceed that of the constraint. Blind activation of upstream resources can quickly inflate stocks.

Rule 7: Transfer batch and Production batch Do Not Need to Be Equal

We call transfert batch, the amount that is transferred from one operation to another.

We call production batch, the quantity produced between two changeover.

A transfer batch lower than the production batch makes it possible to start production in parallel on several resources, but also to transmit the batch more quickly to the next resource and thus to accelerate the flows.

Rule 8: Production batch must be variable and not fixed

Fixed batch sizes affect flexibility and responsiveness. The production is rarely regular, the variation of the sizes of batches is interesting to better follow the demand.

Rule 9: Establish programs taking into account all constraints simultaneously

Planning and scheduling is exclusively based on the bottleneck. He must always work. It must be ensured that the capacities, but also the material availability are coherent on the upstream positions and allow a 100% use of the bottleneck.

In summary



A system contains a multitude of constraints but only one defines the “output” of the system. This constraint tells you what you can provide to the customer.

Use the 5 ToC step to lift the constraint

As long as the constraint is not lifted, your planning must be done according to it.

Utilisez la DBR method of scheduling

Working on an unconstrained one is a waste of time. This will only increase your stock in progress.

Use the ToC Accounting to understand this phenomenon

Just In Time

Theory of Constraints

Main tool


DBR scheduling


Management of the system by the customer’s need.

Control of the system by the bottleneck.

In a word

Both approaches are in the same direction and complementary. One puts more emphasis on customer needs and the other is more focused on the bottleneck.


1 – E. Goldratt (1984) – The Goal

2 – Y. Avraham (2009) – Combining Lean Six Sigma and the theory of constraints to achieve breakthrough performance

3 – F.A Meyer (2012) – La revolution ToC Lean Six Sigma dans les services

4 – C. Gaspoz (2005) – La théorie des contraintes

5 – E. Schragenheim et H.W. Dettmer (2001) – Constraints and JIT: Not Necessarily Cut-Throat Enemies

Share This