How is the bottleneck identified?

By load: the resource whose queue or required hours exceed available capacity by the largest margin in the planning horizon.

Does the bottleneck move?

Always, in high-mix shops. Treating it as fixed is one of the most common scheduling errors.

How do you protect the bottleneck?

Never starve it (maintain a small queue), never overload it (do not promise past its next slot), and protect its setup clustering so changeovers are minimized.

Production Bottleneck — Manufacturing Scheduling Glossary

Definition

A production bottleneck is the one resource — machine, operator skill, fixture, partner — whose available capacity is currently the binding constraint on the factory's total throughput. Goldratt's Theory of Constraints showed that throughput equals bottleneck output: improving non-bottleneck resources does nothing for total output until the bottleneck moves.

Why it matters

In any real factory, the bottleneck moves week to week. It might be the 5-axis Monday, outside plating Wednesday, the senior programmer Friday. Identifying the current bottleneck by load (not by reputation) and protecting its time is the single most leveraged action a planner can take. Every hour reclaimed on the bottleneck is an hour of throughput; every hour reclaimed elsewhere is just WIP.

Common failure mode

The shop "knows" the bottleneck is the 5-axis cell and focuses every improvement on it. This week the actual bottleneck is outside heat treat, but no one is measuring partner queue, so no one acts. Throughput stays flat while the 5-axis cell gets faster.

How Skody approaches it

Skody identifies the current bottleneck from live load, recomputes as conditions shift, and orients the rest of the schedule around protecting its time — never starving it, never overloading it, never breaking its setup pattern.

Bottleneck

The resource whose limited capacity governs the throughput of the whole factory.

Constrained Resource

Any finite resource that limits how much work the factory can complete in a period.

Finite Capacity Scheduling

Scheduling that respects the real, limited capacity of each machine, operator, and tool.

Queue Buildup

The growth of work waiting in front of a resource when arrival rate exceeds processing rate.