Changeover Time, Explained

Changeover time is the gap between "the last good unit of the previous run" and "the first good unit of the next one." Die changes, color and resin transitions, calibration, tooling swaps, purge, and the inspection that confirms the new setup is actually in spec all live inside that gap. On a schedule built in a spreadsheet, it's usually treated as a rounding error. On the floor, it's often the single biggest source of the gap between planned and actual output.

Why it gets underestimated

Most scheduling tools, including a spreadsheet with a formula column, start from a line's nameplate or historical average rate and multiply it by the order quantity. That gives you run time. It says nothing about the setup time between runs, because that number doesn't live on the SKU or the order. It lives in whoever ran the last few changeovers, and it changes depending on what you're changing from and to.

A few examples of the same "changeover" hiding very different costs:

• Color and resin sequence. Going from a light color to a dark one is a five-minute purge. Going dark-to-light on the same line can mean a full purge cycle, a color-check stop, and scrap you can't ship.
• Diameter or profile family. A die change within a compatible tooling family might be 20 minutes. A change that requires a different downstream calibrator or haul-off can be an hour or more, and it needs a specific person available to do it.
• Repeat runs. The same SKU run twice in a row, back to back, can have close to zero changeover. Schedule it as if every transition costs the same and you're either overbooking the week or leaving real capacity on the table.

None of that shows up in "units per hour." It only shows up once you're trying to explain, in a Friday afternoon status meeting, why a week that looked fully booked on Monday finished twelve hours behind.

What to actually track

You don't need a changeover matrix as sophisticated as an automotive plant's to get most of the benefit. What matters is having a real number, attached to the actual transition, not a single plant-wide guess:

1. A default setup time per line, for the common case, most transitions on that line, most days.
2. Named exceptions for the transitions that aren't the common case: dark-to-light color changes, cross-family tooling changes, anything that needs a second person or a specific calibration step.
3. A place those numbers actually live, so the person building next week's schedule isn't reconstructing them from memory or a sticky note on the machine.

That third point is the one that quietly breaks down in a spreadsheet. A formula can hold a flat setup time. It has a much harder time holding "add 45 minutes if the previous run was color X and this one is color Y," attached to the actual sequence of the week, and an even harder time keeping that rule from silently going stale six months later when someone tweaks it in one cell and not the rest.

How this shows up in RunMark

RunMark treats changeover time as a real property of the transition between two scheduled runs, not a flat constant bolted onto a line's throughput. When you place a run after another one, the scheduler reserves the setup window implied by what's actually changing, whether that's resin, color, diameter, or tooling, using the constraints you've configured for that line, the same way it already enforces line-to-SKU capability. That reservation shows up on the schedule itself, as visible time, not as a gap someone has to notice and account for later.

The result isn't a fully automated changeover optimizer. RunMark doesn't claim to sequence your week for you to minimize total setup time. What it does is make sure the time is accounted for before the run is committed, so the plan you hand to the floor is one you can actually hit, and the person building it next week isn't the only one who knows why Tuesday has a 45-minute gap in it.

If you're scheduling extrusion specifically, the industry pages go into the line-capability and throughput side of this in more depth, and the production-capacity calculator lets you test a changeover assumption against a real order before it hits the schedule.