Throughput Accounting and Theory of Constraints
The Theory of Constraints starts from an observation that sounds almost too simple to be useful: a chain is only as strong as its weakest link, and a factory's output is limited entirely by whatever single resource is its tightest bottleneck — improving anything else does not increase what the factory can actually sell.
1. The Theory of Constraints — five focusing steps
The Theory of Constraints (TOC) manages a system through a fixed five-step cycle, always returning to the first step once a constraint is broken, since breaking one constraint simply reveals the next one.
| Step | Action |
|---|---|
| 1. Identify | Find the system's binding constraint (bottleneck) — the resource with the least spare capacity relative to demand on it |
| 2. Exploit | Get the maximum possible output from the constraint as it currently exists, without spending money on it yet — eliminate its idle time, defects and setup delays first |
| 3. Subordinate | Align every other resource in the system to support the constraint's schedule, even if this means deliberately running non-constraint resources below their own maximum capacity |
| 4. Elevate | Only once exploitation is exhausted, invest in genuinely increasing the constraint's capacity (new machinery, overtime, an additional shift) |
| 5. Repeat | Once this constraint is broken, a new constraint will emerge elsewhere in the system — return to Step 1 |
Step 3 is the step most likely to be misunderstood, and it is worth stating plainly: deliberately allowing a non-constraint machine to sit idle some of the time is correct, not wasteful, if running it flat-out would only build up work-in-progress inventory in front of the constraint without increasing what the factory can actually sell.
A factory's throughput is set entirely by the constraint, so extra output from a non-constraint resource beyond what the constraint can absorb produces no additional sales at all.
2. Throughput accounting's definitions
Throughput accounting redefines several familiar terms in a way that deliberately differs from standard costing, and using the standard-costing meaning of these words in a throughput-accounting answer is the most common error in this topic.
Unlike marginal costing's "contribution," throughput treats direct labour as a largely fixed cost in the short run (since a factory's workforce is rarely hired and fired to match each day's exact production volume), so only raw material cost is deducted from sales revenue to arrive at throughput — this is throughput accounting's single sharpest departure from every other costing technique in the syllabus.
The Throughput Accounting Ratio (TA ratio) evaluates whether a product is worth running through the constraint at all:
A TA ratio greater than 1 means the product generates throughput faster than the factory's operating costs accrue, and is worth producing; a ratio below 1 means the product is actually destroying value by consuming scarce constraint time.
Where more than one product competes for the same constrained resource, products should be ranked and produced in descending order of their TA ratio (or equivalently, their return per factory hour), exactly mirroring the limiting-factor ranking rule already used in ordinary marginal costing, but built around throughput rather than contribution.
3. Drum-Buffer-Rope scheduling (brief)
Drum-Buffer-Rope is TOC's practical scheduling method for keeping the whole factory synchronised to its constraint. The drum is the constraint's own production schedule, which sets the pace ("beat") for the entire factory. A buffer — a deliberately held stock of work-in-progress — is placed just ahead of the constraint, protecting it from ever starving for lack of input due to a minor upstream disruption, since any time the constraint sits idle is output the whole factory can never recover.
The rope is the release mechanism that paces material release into the front of the factory to match the drum's rate, specifically to prevent excess work-in-progress inventory from building up ahead of the constraint beyond what the buffer requires.
Worked Examples
Example 1. A factory's single bottleneck machine has 400 available hours per month. Product P requires 2 bottleneck hours per unit and has a selling price of ₹500 with material cost of ₹200 per unit. The factory's total operating (conversion) cost is ₹80,000 per month. Compute Product P's throughput per unit, return per factory hour, and TA ratio.
Throughput per unit = ₹500 − ₹200 = ₹300. Return per factory hour = ₹300 ÷ 2 = ₹150. Cost per factory hour = ₹80,000 ÷ 400 = ₹200. TA Ratio = ₹150 ÷ ₹200 = 0.75.
Example 2. Based on Example 1's TA ratio of 0.75, should the factory continue producing Product P?
A TA ratio below 1 means Product P generates throughput more slowly than the factory's operating cost accrues, so on a pure throughput-accounting basis it is destroying value by consuming scarce bottleneck time — the factory should investigate whether price can be raised, material cost reduced, or bottleneck hours per unit reduced, or consider whether an alternative product with a higher TA ratio should be prioritised for the constraint instead.
Example 3. Two products compete for the same bottleneck resource: Product X has a return per factory hour of ₹180, Product Y has ₹220. If bottleneck capacity is limited, which product should be prioritised, and why?
Product Y should be prioritised, since it generates a higher return per hour of the scarce constrained resource — exactly the same ranking logic used for any limiting-factor decision, applied here to the bottleneck specifically.
Example 4. A factory manager wants to increase output by running a non-bottleneck machine at full capacity, building up extra finished sub-assemblies ahead of the bottleneck "just in case." Using TOC's Step 3 (Subordinate), explain why this could be the wrong decision.
If the non-bottleneck machine's extra output exceeds what the bottleneck can actually process and sell onward, the additional sub-assemblies simply pile up as excess work-in-progress inventory ahead of the constraint, tying up working capital without increasing the factory's actual saleable output — since total throughput is set entirely by the constraint's own capacity, not by how busy the non-constraint machines look.
Example 5. Explain, using throughput accounting's definition, why labour cost is treated differently here than in ordinary marginal costing.
Ordinary marginal costing treats direct labour as a variable cost, deducted from sales along with materials to arrive at contribution. Throughput accounting instead treats labour as a largely fixed cost in the short run (since a workforce is rarely adjusted daily to match exact output), so only material cost — the genuinely "totally variable" cost — is deducted from sales revenue to compute throughput.
Example 6. In Drum-Buffer-Rope scheduling, what is the specific purpose of the buffer placed ahead of the constraint?
To protect the constraint from ever running out of work (starving) due to a minor disruption somewhere upstream, since any idle time at the constraint is lost output the entire factory can never make up.
Example 7. Explain, in TOC's own logic, why "Step 5: Repeat" is a necessary part of the cycle rather than a one-time fix.
Once the current constraint's capacity is elevated (Step 4) enough that it is no longer the tightest limiting resource in the system, some other resource inevitably becomes the new binding constraint, since a system always has exactly one tightest link at any given time — the five-step cycle must therefore return to Step 1 to identify this new constraint, rather than assuming the improvement process is complete after fixing the first one.
Summary
The Theory of Constraints manages a system through five focusing steps — identify, exploit, subordinate, elevate, repeat — treating the single binding constraint as what genuinely limits total output, and correctly allowing non-constraint resources to run below their own maximum capacity where running them further would only build inventory rather than increase actual sales.
Throughput accounting redefines throughput as sales revenue minus totally variable (materials-only) cost, treating labour as largely fixed in the short run — a deliberate departure from ordinary marginal costing — and the Throughput Accounting Ratio (return per factory hour ÷ cost per factory hour) tells a firm whether a product is worth running through its scarce bottleneck resource at all, with products ranked by TA ratio when more than one competes for the same constraint.
Drum-Buffer-Rope scheduling operationalises TOC by pacing the whole factory to the constraint's own schedule (the drum), protecting it from starvation with a strategically placed buffer, and pacing material release (the rope) to prevent excess work-in-progress from building up ahead of it.
