Production scheduling and Theory of Constraints glossary

Plain-language definitions of the terms plant teams meet when they schedule around a bottleneck: constraint, drum-buffer-rope, buffer management, work in process and more.

Bottleneck: The step in a process with the least capacity relative to the work it must do. It limits how much the whole plant can ship: an hour lost there is an hour lost for the plant. More →
Constraint: Anything that limits a system from making more of what it is there to make. It can be a machine, a skill, a supplier, the market, or a policy such as batch-size rules. A bottleneck is one kind of constraint.
Theory of Constraints (TOC): A management approach built on one idea: the output of any system is limited by very few constraints, usually one. Improvement starts by finding that constraint and managing the whole system around it.
Five focusing steps: The improvement cycle of the Theory of Constraints. 1) Identify the constraint. 2) Exploit it, getting the most from what it already has. 3) Subordinate everything else to it. 4) Elevate it by adding capacity if still needed. 5) Repeat, without letting inertia become the new constraint.
Drum-buffer-rope (DBR): A production scheduling method from the Theory of Constraints. The constraint’s schedule sets the pace (drum), time buffers protect it and shipping (buffer), and new work is released only at the drum’s pace (rope). More →
Drum: The constraint’s schedule in drum-buffer-rope. Because the constraint decides output, its schedule effectively is the plant’s schedule. More →
Buffer (time buffer): Protection, measured in time, placed in front of the constraint and before shipping so that normal upstream problems don’t starve it or make orders late. More →
Rope: The release rule in drum-buffer-rope. A job is released a buffer’s length before it is needed, tying the start of new work to the drum’s pace instead of to idle machines. More →
Buffer management: The daily practice of tracking how much of each order’s buffer has been used. Typically green (under a third used) means leave it alone, amber means plan a recovery, and red (over two thirds) means act today. More →
Simplified drum-buffer-rope (S-DBR): A version of DBR for make-to-order plants where the market is usually the constraint. Order due dates act as the drum, one shipping buffer protects delivery, and the load on the most likely internal bottleneck is checked before committing dates. More →
Capacity-constrained resource (CCR): A resource that isn’t the main constraint but can become one if it is scheduled carelessly, for example after a mix change or when its own variation peaks.
Production scheduling: Deciding which jobs run, on which resources, in what order and when, so that orders ship on time with the least waste. Also called manufacturing scheduling. More →
Finite capacity scheduling: Scheduling that respects what each resource can actually do in the time available, rather than assuming infinite capacity as basic MRP planning does.
Order release: The moment a job is allowed onto the shop floor. Controlling release is the simplest lever on work in process and lead time.
Push and pull production: In push production, work starts as soon as material and machines are free. In pull production, work starts only when the downstream system, usually the constraint, signals it is needed. Try both in the dice game. More →
Throughput: In the Theory of Constraints, the rate at which a system turns sales into money: revenue less truly variable costs such as materials. More loosely, the rate of finished output.
Work in process (WIP): Jobs that have been started but not finished. Excess WIP hides problems, ties up cash, and lengthens lead times. More →
Lead time: The time from when an order is released, or received, until it ships.
Little’s law: A queueing rule that holds for stable systems: average work in process = throughput rate × average lead time. Holding output steady, halving WIP halves lead time.
Protective capacity: Spare capacity kept on non-constraint resources so they can catch up after disruptions and keep the constraint fed. Running every machine at full utilisation removes it.
Changeover (setup): The time to switch a resource from one job to the next. At the constraint, every minute of changeover is lost output, so sequencing to cut changeovers there pays off directly.
Full kit: Releasing a job only when everything it needs is ready: material, drawings, tooling and instructions. It stops half-started jobs clogging the floor.
Expediting: Rushing individual jobs past the normal sequence. Constant expediting is a symptom of a schedule that doesn’t reflect the real constraint. More →
High-mix, low-volume (HMLV): Plants that make many different products in small quantities, often make-to-order. Routings vary and the load moves around, so the bottleneck can be hard to see. More →

See the ideas in action.

Run a five-station line with dice and compare push with pull.

Play the dice game

Start with one plant.

A pilot at one site, fixed fee, fixed period. It fits if one step limits your output and your team can give it time.

Tell us about your plant