Reducing Changeover Time Without Buying New Equipment

When changeover time becomes a visible problem, the instinct is often to look at capital: a quick-change die system, a second calibrator, automation that swaps tooling faster. Sometimes that's genuinely the answer. Often, though, a meaningful chunk of changeover time is process, not hardware, and process is free to fix.

Separate what has to happen while the line is stopped from what doesn't

A large share of changeover time on many lines is spent doing things that didn't actually need the line to be stopped: walking to get a tool, hunting for the right die, waiting on a material delivery that could have arrived earlier. The classic lean manufacturing distinction here is internal setup (work that genuinely requires the line to be down) versus external setup (work that can happen while the previous run is still going).

Moving as much as possible into the external category, staging the next die, color, or material at the line before the current run finishes, doesn't require new equipment. It requires deciding in advance what's coming next and having it ready, which is a scheduling and staging discipline, not a purchase.

Sequence to avoid the expensive transitions

Not every changeover costs the same. A light-to-dark color change might be a five-minute purge. The reverse can mean a full purge cycle and a color-check stop. A same-family tooling change might be twenty minutes. A cross-family change might need an hour and a specific person available to do it.

If the order in which jobs run is somewhat flexible, sequencing similar setups next to each other, grouping colors light to dark, keeping tooling families together where due dates allow it, avoids paying for the expensive transitions more often than necessary. This only works if you actually know which transitions are expensive, which means tracking real changeover time per transition rather than a single average. You can't sequence around a cost you haven't measured.

Standardize the procedure itself

Ask five people how they do the same changeover and you'll often get five slightly different answers, with different times attached. Some of that variation is unavoidable. A lot of it is just nobody having written down and timed the actual steps. Documenting the real sequence, timing each step, and identifying which parts are genuinely fixed (a calibration that physically takes X minutes) versus which parts vary by who's doing it (searching for a tool, deciding what order to do things in) turns a fuzzy target into something you can actually shrink. Often the biggest win isn't cutting the average time, it's cutting the variance, so a changeover reliably takes 25 minutes instead of sometimes 15 and sometimes 50 depending on who's running it and how organized they were that day.

Stage tools and materials where they're actually needed

If the tools, dies, or gauges for a changeover live in a cabinet three aisles away, every changeover pays a walking tax on top of the actual work. A staging area or cart near the line, prepped ahead of time with what the next changeover needs, removes that tax without buying anything beyond a cart and some organization.

Cross-train so a changeover doesn't wait on one specific person

If only one person on a shift knows how to do a particular changeover, that changeover's real duration includes however long it takes to find that person and get them to the line, which has nothing to do with the mechanical work itself. Training a second person to do the same changeover removes a queuing delay, not a technical one, and it's often the single cheapest fix on this whole list.

You can't fix what you don't measure

Every lever above depends on the same prerequisite: knowing the real changeover time for each type of transition, not a flat plant-wide guess. Without that, you can't tell whether sequencing changes actually helped, whether standardizing a procedure reduced variance, or whether a new staging cart made a measurable difference. Measurement isn't the fix by itself, but it's what tells you whether any of the fixes above actually worked.

Where RunMark fits

RunMark tracks changeover time as a real, attached property of each scheduled transition rather than a flat constant, and its scheduler factors that real cost into how it sequences the week. That gives you two things a spreadsheet-based schedule usually can't: a genuine before-and-after number to measure whether a process change actually reduced changeover time, and a scheduler that naturally favors a sequence that avoids expensive transitions when the order mix allows it, rather than leaving sequencing entirely to guesswork. None of that replaces doing the process work itself, staging, standardizing, cross-training, but it does mean the improvements are measurable instead of a feeling that things got better.