Theory of Constraints Explained: The Five Focusing Steps

The Theory of Constraints is the conceptual foundation behind throughput accounting, and it's a distinct topic in its own right on ACCA Performance Management and CIMA's management accounting syllabus. Understanding the theory itself — not just the throughput accounting calculations built on top of it — is what lets students correctly interpret scenario-based exam questions about bottlenecks and constraint management.

What is the Theory of Constraints?

The Theory of Constraints (TOC) was developed by Eliyahu Goldratt and popularised through his bestselling business novel The Goal. Its central idea is that every system — a factory, a supply chain, any process with multiple linked steps — has at least one constraint (or "bottleneck") that limits the overall output of the entire system, no matter how efficient every other part of the process is. As Goldratt put it, strengthening any link in a chain other than the weakest one is essentially wasted effort, because the chain's overall strength is still determined by its weakest point.

The five focusing steps

TOC provides a structured, five-step process for managing constraints and driving continuous improvement:

• 1. Identify the constraint — determine which specific resource or step is actually limiting the system's overall output. This isn't always obvious, and misidentifying the real constraint leads to improvement efforts being wasted elsewhere.
• 2. Exploit the constraint — get the maximum possible output from the existing constraint before spending money on additional capacity. This might mean eliminating idle time on the bottleneck machine, or ensuring it's never starved of work.
• 3. Subordinate everything else to the constraint — align every other part of the process to support the constraint's output, rather than optimising each step in isolation. Non-constraint resources should produce only as much as the constraint can actually use, to avoid a build-up of unnecessary work-in-progress inventory.
• 4. Elevate the constraint — once the first three steps are exhausted, invest in additional capacity at the constraint itself, such as new equipment or extra staff, to genuinely increase what the system can produce.
• 5. Prevent inertia and repeat — once a constraint is elevated, the bottleneck typically shifts somewhere else in the system. Return to step one and repeat the cycle, rather than assuming the improvement process is finished.

Drum-buffer-rope scheduling

TOC's practical scheduling method is often called drum-buffer-rope. The "drum" is the constraint itself, which sets the pace or "beat" for the entire production process — nothing can move through the system faster than the constraint allows. The "buffer" is a strategically placed stock of work-in-progress positioned just ahead of the constraint, protecting it from ever running out of work due to minor disruptions elsewhere. The "rope" is the signal that paces the release of new work into the system, timed to match what the constraint can actually process — releasing work faster than the constraint can handle it just builds up unnecessary inventory without increasing overall output.

Why this connects directly to throughput accounting

TOC's central philosophy — that the constraint determines everything — is exactly why throughput accounting treats labour and overhead costs as largely fixed in the short run and focuses management attention specifically on maximising throughput at the bottleneck. Investment (money tied up in inventory) is treated cautiously in TOC thinking, since inventory sitting anywhere except immediately before the constraint doesn't help overall output and just ties up cash that could otherwise be used productively.

Why treating every process as equally important is a mistake

A common intuitive error is trying to improve every step of a process by the same amount, assuming this will proportionally improve overall output. TOC's insight is that improving a non-constraint resource does nothing for overall system output — a faster machine upstream of the bottleneck just produces more inventory that piles up waiting for the bottleneck to catch up. All meaningful improvement in total output has to come through the constraint itself, which is precisely why identifying the real constraint correctly, as step one, is so important.

FAQs

Can a system have more than one constraint at the same time? In TOC theory, a well-designed system typically has one dominant constraint at any given time; if genuinely multiple resources are equally limiting, TOC still recommends focusing improvement effort systematically, one constraint at a time, rather than trying to elevate everything simultaneously.

Does elevating the constraint always solve the problem permanently? No — once a constraint is elevated enough, the bottleneck typically moves to a different part of the system, which is why the fifth focusing step explicitly instructs returning to step one rather than treating the process as finished.

How is the Theory of Constraints different from lean/Just-In-Time thinking? The two share some common ground on minimising unnecessary inventory, but TOC specifically focuses improvement effort on the single limiting constraint, while JIT more broadly aims to eliminate waste and excess inventory across the whole production process.

The Theory of Constraints reframes process improvement around a simple but often-overlooked truth: a system can only ever be as productive as its most limiting step, so genuine improvement has to start there, not everywhere at once.