When a bottleneck moves, the project did not fail

Production teams sometimes treat a moving bottleneck as a modelling problem. They improve station 4, the rate-limiter becomes station 7, and the question becomes whether the first analysis was wrong. In many cases it was right. The line changed because the first constraint was relieved.

A useful bottleneck map does not present a permanent villain. It shows the current rate-limiter, the contribution to cycle-time loss and the next likely station after the first improvement. That sequence is more useful than a single ranking because it supports staged work rather than one large, unfocused programme.

The practical step is to review bottleneck rank by shift and product mix. If the same station dominates in every view, the case is straightforward. If the rank changes by mix, operations may need a planning adjustment rather than a machine change.