Theory of Constraints

← All maturity guides

SEF Reference Guide G11  |  Manufacturing OpEx Maturity Assessment

What it is

Every plant has one resource that limits how much it can ship. It might be a machine, a test cell, a skilled welder, an engineering group, or a supplier. The Theory of Constraints (TOC) says you find that one resource, make it work as hard as possible on the right things, and set the pace of everything else by it. Output of the whole plant equals output of the constraint, so an hour gained at the constraint is an hour gained for the business, and an hour gained anywhere else changes nothing.

TOC comes from Eliyahu Goldratt, who brought it to a wide audience through the business novel The Goal (1984, written with Jeff Cox). The practical core is the Five Focusing Steps, a repeating cycle for finding and managing the constraint.

TOC and lean fit together. Lean removes waste everywhere. TOC tells you where removing waste will actually raise output first. Most plants get the best results by pointing their lean tools (SMED, TPM, standard work, problem solving) at the constraint before anywhere else.

Why it matters

• Delivery: the constraint sets lead time and on-time delivery. Protecting it is the fastest way to stop late orders.
• Cost and profit: extra output from the constraint is sold with almost no added cost, so most of it drops to the bottom line.
• Cash: releasing work at the pace of the constraint stops WIP piling up in front of it, which frees cash tied up in inventory.
• Capital: you often find the capacity you need by exploiting the constraint before buying a second machine.
• People: a clear focus ends the argument over which improvement matters most. Everyone can see why the constraint comes first.

Key terms

Term	Meaning
Constraint	The operation or resource where demand exceeds capacity. It sets the output of the whole system. Also called the bottleneck.
TOC	Theory of Constraints (Eliyahu Goldratt). Every system’s output is limited by at least one constraint, and managing that constraint is the fastest path to more throughput and profit.
Five Focusing Steps	The TOC improvement cycle: 1) Identify the constraint, 2) Exploit it, 3) Subordinate everything else to it, 4) Elevate it, 5) Repeat, and do not let inertia become the new constraint.
Throughput (T)	In TOC accounting, the rate at which the business generates money through sales: sales revenue minus truly variable costs (mainly raw material). Units made but not sold are not throughput.
Inventory / Investment (I)	Money tied up in things the business intends to sell, plus equipment and facilities.
Operating Expense (OE)	Money spent to turn inventory into throughput: wages, utilities, overhead.
Drum-Buffer-Rope (DBR)	TOC scheduling. The constraint is the drum that sets the pace, a time buffer of work protects it from starving, and the rope ties material release to the drum’s pace so WIP stays under control.
Buffer management	Watching how much of the constraint’s buffer has been consumed (often in green, yellow, and red zones) and acting on red.
Market constraint	The case where the plant can make more than customers will buy. The constraint is then in sales, not on the floor.
Policy constraint	A rule or habit (batch sizes, efficiency targets, shift patterns) that limits output more than any machine does.

How to do it

1. Map the flow and collect load data. For each major step, compare required hours (demand x standard time) with available hours for the next 4 to 8 weeks. The step with the highest load percentage is your candidate constraint. The SEF Resource Calculator (see G26) does this check for people and machines, including batch equipment.
2. Confirm on the floor. The constraint usually has the largest pile of WIP in front of it, the most overtime, and people downstream waiting for it. If the data and the floor disagree, trust the floor and fix the data.
3. Check for a market or policy constraint. If every step has spare capacity and you still miss orders, the limit is in sales, planning, or a policy. Name it honestly.
4. Exploit the constraint. Get more good output from what you already have:
   • Keep it running through breaks, lunches, and shift changes (staggered relief).
   • Stage the next job, tooling, and material before the current job ends.
   • Apply SMED to its changeovers first and TPM to its breakdowns first.
   • Inspect before the constraint so it never works on bad parts.
   • Run only work that has a real customer order behind it.
5. Subordinate everything else. Release material into the line at the constraint’s pace (the rope). Stop rewarding non-constraint areas for running ahead; their spare capacity is there to protect the constraint. Size a time buffer in front of the constraint and track its color every shift.
6. Elevate only after you exploit. Add capacity (overtime, a second shift, outsourcing, a new machine) when exploitation and subordination are done and the constraint still limits sales.
7. Repeat. When the constraint moves, go back to step 1. Update your schedules, buffers, and measures so old rules do not hold the new constraint back.
8. Measure throughput. Track constraint output per shift, constraint lost time by reason, buffer penetration, and plant throughput (shipped value minus material) week by week.

What each maturity level looks like

FL-1 Constraint Management (Theory of Constraints)

Level	Assessment wording	What it looks like on the floor	How to move up
1	No evidence that constraints have been identified.	Different managers name different bottlenecks. Every area pushes to be busy. WIP piles up in several places. Overtime is spread everywhere.	Run a load versus capacity check on the main steps and walk the floor to confirm where WIP collects. Agree on one named constraint.
2	The constraint is known and managed ad hoc.	People can name the bottleneck, and it gets attention when orders are late. It still stops for breaks, waits for material, and runs jobs with no order behind them.	Put the constraint on the daily meeting board. Track its lost time by reason every shift. Start exploit actions (staggered breaks, staging, quality check before it).
3	As Level 2, plus improvement is systematically focused on the constraint.	SMED, TPM, and problem solving projects are aimed at the constraint first. Its lost time Pareto is posted and shrinking.	Formalize the Five Focusing Steps: add subordination (release at the constraint’s pace, time buffer) and a clear rule for when to elevate.
4	As Level 3, plus the Five Focusing Steps of TOC drive improvement.	Material release is tied to the constraint (DBR). Buffer status is reviewed every shift. Non-constraint areas are not measured on utilization. The team re-checks where the constraint is after each big change.	Build a throughput trend (constraint output and plant throughput) and review it monthly with leadership. Show the link between constraint actions and results.
5	As Level 4, plus clear evidence of improving throughput.	A 12-month trend shows rising constraint output and plant throughput, with WIP and lead time flat or falling. When the constraint moved, the team caught it and started again.	Sustain: keep the cycle running, share the method with other sites, and extend it to planning, engineering, and supply constraints.

Evidence to look for

Document

• Load versus capacity analysis naming the constraint, with date.
• Constraint lost-time log with reason codes.
• Scheduling or release rules that reference the constraint (DBR, CONWIP, or equivalent).
• Improvement project list showing which projects target the constraint.

Observation

• Is the constraint running during breaks and shift changes?
• Is the next job staged and ready before the current one ends?
• Where does WIP collect? Is there a planned buffer in front of the constraint, and is WIP elsewhere small?
• Is there a board at the constraint showing output versus plan and buffer status?

Statement (questions to ask and who to ask)

• Operators and supervisors: “Which operation limits how much we ship? How do you know?”
• Planner or scheduler: “How do you decide when to release a job to the floor?”
• Maintenance lead: “Does the constraint get priority for repairs and PM?”
• Plant manager: “When did the constraint last move, and what did you change?”

Management report

• Weekly constraint output versus plan, with lost-time Pareto.
• Throughput trend (sales minus truly variable cost) over 12 months.
• Buffer penetration summary (how often red, and top causes).
• OTD and lead time trends alongside constraint output.

Common pitfalls

• Improving the wrong step. Cutting cycle time at a non-constraint looks good on a report and does not raise output. Goldratt’s line: an hour saved at a non-bottleneck is a mirage.
• Measuring everyone on utilization. If every area is judged on being busy, they will run ahead of the constraint, build WIP, and hide the real problem.
• Buying capacity first. Spending capital before exploiting and subordinating often buys a machine that was not needed, or just moves the pile of WIP.
• Starving or blocking the constraint. No buffer in front of it, or no space after it, and the constraint stops for reasons that have nothing to do with its own capability.
• Letting quality fail after the constraint. Scrapping a part downstream throws away constraint time. Put the tightest quality checks just before and just after it.
• Forgetting step 5. Once the constraint moves, old rules keep favoring the old bottleneck. Inertia becomes the new constraint.

Template

Constraint identification and exploitation sheet

Step / resource	Weekly demand (units)	Std time per unit (h)	Required hours	Available hours	Load %	WIP in front (units)	Constraint? (Y/N)

Load % = Required hours / Available hours x 100. Highest load % (confirmed by WIP and floor observation) is the constraint.

Constraint daily log

Date	Shift	Planned output	Actual output	Lost minutes	Top loss reason	Buffer status (G/Y/R)	Action and owner

Five Focusing Steps checklist

• [ ] Constraint named and agreed by operations, planning, and maintenance
• [ ] Exploit: breaks covered, next job staged, quality check before constraint, SMED and TPM started
• [ ] Subordinate: release rule tied to constraint pace, time buffer sized, utilization targets removed elsewhere
• [ ] Elevate: capacity options costed and approved only after exploit and subordinate
• [ ] Repeat: constraint location re-checked monthly and after major changes

First 30 days

• Week 1: Build the load versus capacity sheet for the main value stream. Walk the floor and note where WIP piles up. Agree on the constraint with the plant leadership team.
• Week 2: Start the constraint daily log. Add the constraint to the daily meeting board. Cover breaks and shift changes at the constraint.
• Week 3: Build the lost-time Pareto. Launch one SMED or TPM action on the top loss. Move a quality check to just before the constraint.
• Week 4: Write a simple release rule tied to constraint output and size a starting time buffer. Review results: constraint output per shift versus week 1.

Further reading

• Eliyahu M. Goldratt and Jeff Cox, The Goal: A Process of Ongoing Improvement (North River Press, 1984; later revised editions).
• Eliyahu M. Goldratt and Robert E. Fox, The Race (North River Press, 1986). Covers Drum-Buffer-Rope scheduling.
• Eliyahu M. Goldratt, What Is This Thing Called Theory of Constraints and How Should It Be Implemented? (North River Press, 1990).
• James F. Cox III and John G. Schleier Jr. (eds.), Theory of Constraints Handbook (McGraw-Hill, 2010).
• Theory of Constraints International Certification Organization (TOCICO): https://www.tocico.org