Theory of Constraints

“Tell me how you measure me and I will tell you how I will behave.” — Goldratt

Theory of Constraints (TOC) is a management philosophy developed by Eliyahu Goldratt, introduced in his 1984 business novel The Goal. Its central claim: every system has exactly one constraint that limits its throughput at any given moment. All improvement effort that doesn’t address the constraint is waste.

---

The Goal

The goal of a for-profit organization is to make money now and in the future. Goldratt defines three operating metrics that replace cost accounting:

Metric	Definition	What it Captures
Throughput (T)	Revenue minus totally variable costs	Rate at which the system generates money through sales
Inventory (I)	Money invested in things intended to be sold	All tied-up capital: raw material, WIP, finished goods, equipment
Operating Expense (OE)	Money spent to turn inventory into throughput	Labor, overhead, fixed costs

Profit = T − OE

The power of this framework: increasing T is multiplicative; reducing OE is bounded at zero. Goldratt argued that managers spend 90% of their attention on reducing OE and almost none on increasing T — which is why they sub-optimize.

---

The Five Focusing Steps (POOGI)

POOGI = Process of On-Going Improvement. The five steps form a loop that never ends.

1. Identify the Constraint

Find the one resource, step, or policy limiting system throughput. In a warehouse: the constraint might be the packing line, a narrow dock window, a single piece of equipment that everything flows through, or an information bottleneck (WMS queue, manual manifesting).

Herbie heuristic: In The Goal, a slow hiker named Herbie constrains the entire troop. Inventory piles up behind him; everyone ahead of him is idle. Find the Herbie in your operation.

2. Exploit the Constraint

Get maximum output from the constraint without spending money. The constraint must never wait — it must never be starved of work or blocked by downstream bottlenecks.

Examples in a DC:

• Eliminate breaks at the pack station by staggering operator schedules
• Remove non-value-added steps from the constraint (move QC checks upstream)
• Dedicate the most experienced operators to the constraint workstation
• Pre-stage materials so the constraint never waits for inputs

3. Subordinate Everything Else to the Constraint

All other resources must pace themselves to the constraint’s rate. Running non-constraint resources at their maximum only builds inventory ahead of the constraint and creates chaos downstream.

This is counter-intuitive: idle time at a non-constraint is not waste — it is correct behavior if the constraint is not yet ready.

4. Elevate the Constraint

If the constraint is still limiting performance after exploitation, invest to increase its capacity. Examples: add a second packing station, invest in conveyor merge capacity, add sorter lanes.

Only elevate after exploiting — throwing capital at a poorly-run constraint is expensive and often doesn’t solve the problem.

5. Return to Step 1 — Do Not Let Inertia Become the Constraint

When the constraint moves (and it will), the rules and policies built around the old constraint become inertia. Identify the new constraint and repeat.

---

Drum-Buffer-Rope (DBR)

DBR is TOC applied to production/warehouse scheduling:

• Drum: The constraint sets the pace. The entire system produces at the drum’s rate.
• Buffer: Protective inventory staged ahead of the constraint ensures it is never starved. Buffer size is set by variability, not average demand.
• Rope: A signal that limits the release of new work into the system to match the drum rate. Prevents WIP accumulation ahead of non-constraints.

DBR in a DC: the shipping sorter (drum) runs at a fixed rate; the buffer is a queue of packed cartons staged before the sorter; the rope releases orders from picking only fast enough to maintain that buffer.

---

TOC vs. Lean vs. Six Sigma

Lens	Target	Tool	Limitation
TOC	The constraint	Five Focusing Steps, DBR	Ignores waste at non-constraints (by design)
Lean	All waste everywhere	Value stream mapping, kaizen	Can optimize non-constraints and miss the bottleneck
Six Sigma	Variation	DMAIC, SPC	Reduces variation but may not address capacity

Most effective: TOC identifies where to focus → Lean eliminates waste at the constraint → Six Sigma reduces variation at the constraint. Run them in that order.

---

Statistical Fluctuations and Dependent Events

Goldratt’s foundational insight before the five steps: in a system of dependent events (each step requires the prior step’s output), statistical fluctuations accumulate — they do not average out.

Example: if each of 10 steps in a pick-pack-ship sequence has 90% on-time performance, the end-to-end on-time rate is not 90% — it is 0.9^10 = 35%. This is why individual-step metrics look fine but the warehouse misses ship windows.

Implication: buffer the constraint, not the end of the line.

---

Throughput Accounting vs. Cost Accounting

Traditional cost accounting treats idle labor and capacity as waste → managers reduce headcount → the constraint gets starved → throughput drops → decisions made to “cut costs” destroy the ability to ship.

TOC throughput accounting:

• Idle time at a non-constraint = free. It costs nothing extra.
• Idle time at the constraint = lost throughput. It costs T per minute.
• Correct decision: invest in the constraint; let non-constraints idle.

---

Applications in DC and Warehouse Operations

DC Function	Typical Constraint Candidates
Receiving	Dock door count, LPN scanning rate, putaway transport capacity
Pick	Replenishment rate to forward pick, travel distance, single-SKU pick faces
Pack	Pack station count, cartonization time, void-fill throughput
Ship	Sorter rate, manifesting speed, dock door availability vs. carrier arrival pattern
Returns	Disposition decision throughput (human grading), tote transport from dock to grading

Finding the constraint in a DC: Build the throughput model for each function. The step with the highest utilization (closest to 100%) at Design Day volume is the constraint. Validate by observing where inventory queues pile up during peak hours.

---

See Also

• Throughput Analysis — volume modeling, design day calculation, peak-to-average ratios
• Lean Methods in DC Operations — waste elimination framework
• Six Sigma DMAIC in DC — variation reduction at the constraint
• CI Program Foundation — organizational infrastructure for ongoing improvement
• 5S and Visual Management — making the constraint and its status visible in real time