How Setup Reduction Funds the Overtime You Avoid

Every changeover hour that better sequencing recovers is a production hour handed back to the same shift, and a production hour handed back is an overtime hour you never have to buy. Much of the overtime shops run routinely is capacity that setup quietly stole during the week, and sequencing takes it back for free. EDGEBIC by User Solutions cuts changeover by scheduling jobs in an order that avoids the expensive transitions, which turns a shift that was overflowing into one that fits, and removes the reason the overtime existed.

This post is about one specific swap: setup hours recovered funding overtime avoided. It sits under the EDGEBIC results guide and connects two mechanisms already documented, how EDGEBIC cuts changeover hours and how EDGEBIC reduces unplanned overtime. For the overtime lever itself, see what an overtime allowance is in shift scheduling.

Overtime Is Often Capacity Theft Wearing a Payroll Disguise

Shops run overtime for two very different reasons that look identical on the payroll report. The first is genuine excess demand: real orders that exceed real capacity, where overtime is the honest tool. The second is recovery: the shift ran out of hours because setup ate them, so you buy overtime to catch up on work that should have fit.

The second kind is capacity theft in disguise. A machine that spends three hours changing over between jobs has three fewer hours to make parts, and the parts that changeover displaced do not disappear. They pile into overtime or spill past the ship date. The overtime looks like a demand problem on the report, but its cause is a sequence problem on the floor.

The reason this matters for ROI is that the two kinds of overtime have completely different fixes. You cannot sequence your way out of genuine excess demand. You can absolutely sequence your way out of overtime that only exists because changeover stole the capacity, and that second kind is a large share of most shops' catch-up hours.

The Mechanism: The Same Jobs, a Better Order

Changeover on many machines is sequence-dependent: the time depends on what ran before. EDGEBIC's setup matrix charges the true from-to time and lets the scheduler cluster jobs to avoid the expensive transitions.

The documented paint booth case makes the swap concrete. Three jobs on one booth, where white to black costs 60 minutes and black back to white costs a 240-minute solvent purge:

The 8.75-hour day overflows the shift, which is exactly the condition that forces overtime or a spilled job. The resequenced day is 4.75 hours and fits comfortably. Nothing changed but the order of the jobs: no new equipment, no faster paint, no overtime. The full walkthrough traces every minute.

That is the swap in its simplest form. The overflow that would have been covered by overtime is erased by sequencing, so the overtime is not avoided by working harder. It is avoided because the work fits inside regular hours once the changeover is not stealing them.

Why This Is the Self-Funding Version

Most efficiency improvements cost something to get: capital for a faster machine, capital for a second one, training for a new process. Sequencing costs almost nothing once the changeover times are loaded as data, because it recovers capacity from a resource you already own by running it in a smarter order.

That is what makes it self-funding rather than an investment. The recovered changeover hours are capacity delivered at close to zero marginal cost, and they land exactly where overtime would otherwise be spent. You are not trading dollars for hours. You are converting a sequencing decision, which is free, into capacity that displaces overtime, which is expensive. The saving is the overtime premium on every hour the sequencing hands back.

At scale the effect compounds. The paint case was three jobs on one booth. A machine running a full queue every day with a worse mix has more expensive transitions to avoid, and the same clustering logic applies every single day. Once the from-to times are data instead of tribal knowledge, the size of the recoverable capacity, and therefore the avoidable overtime, is computable before you chase it.

Where the Swap Pays Most: the Constraint

The swap is worth the most on your bottleneck, because setup hours there are throughput hours, and throughput hours lost are what overtime is usually bought to recover.

On a non-constraint, recovering four changeover hours is a convenience: the machine had slack anyway. On the constraint, those four hours are four hours of additional plant output, and if the plant was going to buy overtime somewhere to make its numbers, the constraint is where that overtime would have paid off. Sequencing the constraint first turns its stolen changeover back into output, which is the output the overtime was going to chase.

So the priority is clear: sequence the constraint before anything else, because that is where a recovered setup hour most directly cancels an overtime hour. Sequencing an idle machine cancels no overtime, because that machine was never the reason you ran late.

Sizing the Swap

The saving is the overtime premium on the hours sequencing recovers. It is measurable with a clipboard.

1. Log changeover on your worst one or two machines for two weeks. Actual minutes, next to the machine. This is the number the matrix moves and the one shops rarely measure separately.
2. Log the overtime authorized in the same window, and why. Separate genuine excess demand from catch-up overtime.
3. Estimate the recoverable share. Using your real changeover matrix, how much of the logged setup is avoidable through clustering? The paint case's 73% is the shape, not a promise; your number depends on your mix and your worst transition.
4. Price it at the overtime premium. The recovered hours displace overtime, so value them at the premium over straight time, not just the base rate, because the premium is the part you stop paying.

The heritage record shows what recovered capacity is worth at a real shop: Technical Glass Products accommodated a 4% increase in business with its existing workforce, which is demand absorbed without new labor. Capacity handed back by smarter scheduling is capacity you do not have to buy in overtime or headcount.

What the Swap Cannot Do Alone

Sequencing recovers avoidable changeover. It cannot manufacture capacity beyond that, and it should not be sold as if it can.

It cannot cover genuine excess demand. If real orders exceed real capacity even after every changeover is optimized, overtime is the correct tool and the swap is exhausted. Sequencing removes the avoidable overtime, not the necessary kind.

It cannot beat a matrix full of guesses. The 73% recovery exists because the changeover times were real. If the from-to values are estimates, the sequencing optimizes fiction and the recovered hours are fiction too. Load true times, verified against logged actuals, or the swap is imaginary.

It cannot reduce setup that is not sequence-dependent. Where every changeover takes the same time regardless of order, clustering has nothing to exploit, and the answer is setup reduction engineering, not sequencing. The swap needs a matrix with expensive transitions to avoid.

It cannot force a due-date-driven sequence to be optimal. Sometimes the customer dates require an order that is not the lowest-setup one. The scheduler shows you the trade, but honoring a hard due date can mean accepting some changeover you would rather cluster away. That is a real constraint, not a failure.

It cannot decide your overtime policy. Whether to bank the recovered capacity as reduced overtime, faster delivery, or more throughput is a management choice. The schedule hands back the hours; what you do with them is yours.

The through-line: a lot of routine overtime is not demand, it is changeover stealing capacity during the week, and sequencing takes that capacity back for free. Recovered setup hours land exactly where overtime would have been spent, which is why the swap funds itself. Want to know how much of your overtime is really a setup problem? Bring two weeks of changeover and overtime logs to a demo, and we will run your jobs both ways.