Definition
A bottleneck (or constraint) is a point of congestion in a system that occurs when workloads arrive too quickly for the system to handle. According to the Theory of Constraints, the total throughput of any system is limited by its narrowest point—the bottleneck. Consequently, any improvement made at a non-bottleneck point is a waste of resources.
Why It Matters
According to the Theory of Constraints, any effort spent optimizing a non-bottleneck is a total waste; identifying the true choke point is the only way to achieve meaningful increases in system throughput.
Core Concepts
-
Theory of Constraints (TOC): A management philosophy introduced by Eliyahu M. Goldratt. It posits that every system has at least one constraint; focusing on this point provides the highest leverage for improving system performance.
-
The Choke Point Principle: If you have a process with steps A, B, and C, and step B is the bottleneck, increasing the capacity of A or C will only lead to a larger “pile” of work waiting at B. Total system speed is .
- How to read: “The total system speed equals the minimum of the speeds of A, B, and C.”
- Meaning: System throughput is capped by the slowest step—optimizing non-bottlenecks yields zero net gain.
-
Moving Bottlenecks: Solving a bottleneck does not eliminate constraints entirely; it merely shifts them to the next narrowest point in the system.
-
Hidden Bottlenecks: Constraints that are not immediately obvious, such as distribution in a startup (Thiel’s “Dead Zone”), power availability in data centers (Nvidia), or human decision-making speed in an automated workflow.
-
The Great Filter: In evolutionary biology and cosmology, a hypothetical bottleneck that prevents life from becoming a spacefaring civilization.