Theory of constraints

What it means

The theory of constraints is a management approach holding that the output of a whole system is set by its weakest point, the so-called constraint or bottleneck. Improvements elsewhere therefore will not raise the overall result until that constraint is dealt with.

Its author was the Israeli physicist Eliyahu Goldratt, who popularised the idea in his novel The Goal. He compared a company to a chain that is only as strong as its weakest link. Strengthening the other links costs money but does not make the chain any stronger.

A constraint may be a machine, a worker, a department, a supplier, or even a company rule or way of thinking. The theory offers a five-step process: identify the constraint, exploit it as fully as possible, subordinate everything else to it, elevate it and then look for the next one.

How it works in practice

In manufacturing, a bottleneck often shows as a pile-up of work in progress in front of it and idle time behind it. Value stream mapping helps reveal where the flow of work stops, and capacity planning shows which part is overloaded.

In services, the constraint may be a single experienced expert who must approve every tax matter. The solution is to relieve them of routine tasks, train others or redesign the process so that they review only what genuinely needs their expertise.

The theory of constraints complements lean and Six Sigma well. While lean removes waste across the whole process and Six Sigma reduces variation, the theory of constraints helps decide where to start so that improvement has the greatest effect on the overall result.

What to watch out for

Once one constraint is removed, the bottleneck moves elsewhere. It is a continuous process, not a one-off task. A company should regularly check what is holding it back now, otherwise it will keep improving areas that no longer decide the outcome.

Traditional measures such as the utilisation of each machine or worker can lead to poor decisions. Trying to keep everyone fully busy often creates inventory and waiting. According to the theory of constraints, throughput of the whole system matters more than local efficiency.