Lean Manufacturing

SMED Changeover Reduction: Converting Internal to External Setup Time

Every minute of changeover is lost production time. Here is how to calculate changeover cost and build the case for SMED work.

Changeover cost = changeover time in hours x (machine hourly rate + opportunity cost per hour), and that formula usually shows bigger losses than people expect. If a press takes 3.5 hours to change over and the machine rate is $85 per hour with $120 per hour of missed contribution margin, each changeover costs 3.5 x ($85 + $120) = $717.50. At 3 changeovers per week for 50 weeks, that is more than $107,000 per year. This is the number that turns setup reduction from a lean slogan into a business case. Without the cost math, SMED efforts often get treated like housekeeping instead of capacity work.

SMED works by separating internal elements, which require the machine to be stopped, from external elements that can be done while it still runs. Typical external candidates are tooling staging, material verification, paperwork, preheating dies, and presetting fasteners or locators. The biggest quick wins usually come from staging tools before shutdown, switching to quick-release hardware, and standardizing clamp heights, shims, or datum surfaces across product families. Plants commonly cut 30% to 50% of changeover time with these basic steps before spending serious capital. Good input data comes from a timed video study, first-good-part definition, and repeat observations across multiple crews, not from one optimistic stopwatch sample.

The most common mistake is measuring only wrench time and ignoring the rest of the lost production window. The honest standard is last good part of the previous run to first good part of the next run, including warm-up scrap, first-piece approval, and parameter stabilization. Teams also claim success after a kaizen event but never lock the new method into standard work, shadow boards, carts, and material call timing. Another miss is treating every setup the same when a few product families drive most of the loss. If you do not know which changeovers happen most often and cost the most, you can spend effort on the wrong setup.

Use the result to decide where SMED will free the most bottleneck hours and whether the change should be procedural or capital. A 75% setup reduction can also justify smaller lot sizes, which cuts WIP and finished goods inventory on top of the direct capacity gain. If lot size drops from 5,000 to 1,000 units, average WIP can fall by 4,000 units for that product family, which may be worth more than the labor saved during the setup itself. The calculation also helps planners load the schedule more realistically and see which machine families need protected setup windows. Shorter changeovers give the schedule more flexibility without buying another machine.

Advanced teams connect setup time to OEE, lot size economics, and inventory turns so the gains do not disappear into the next planning cycle. Some setups are worth reducing because they hit the constraint, while others are not urgent because they happen on non-bottleneck equipment. Track median changeover time, variation by crew, and first-pass success after setup so the improvement is durable. If the new method reduces average setup but increases startup scrap, the real gain may be smaller than the headline time suggests. SMED works best when time, quality, and scheduling are measured together.

Published 2026-05-28.