Smaller
Is Almost Always Better
EPE
Every Product Every ___
WIP
Proportional to Batch Size
SMED
Enables Small Batches

The Batch Size Trap

Large batches feel efficient. You set up the machine once and run a big quantity — maximum utilization, minimum changeovers. The spreadsheet looks great. But large batches create enormous hidden costs that do not show up in the setup-time calculation:

Hidden Cost of Large BatchesHow
Long lead timesA batch of 1,000 takes 10x longer to complete than a batch of 100. Customer waits 10x longer.
High WIPLarge batches = large piles between steps. Cash tied up, space consumed, problems hidden. Little's Law.
Late defect discoveryIf a defect starts at unit 50, you do not find it until unit 1,000 is done. 950 units potentially affected.
InflexibilityCustomer wants Product B? Too bad, you are running a 3-day batch of Product A.
Forecasting dependencyLarge batches require forecasting what to build. Small batches let you respond to actual demand.
Obsolescence riskLarge finished goods batches may not sell before engineering changes or demand shifts.
Transfer batch, drawn Same three machines, same one unit a minute, same thirty-six minutes of work at every step. Moving the work in threes instead of all twelve halves the lead time and puts the first units in the customer's hands four times sooner.

Twelve units, three steps, one minute per unit per step. Whole batch: cut 0–12, machine 12–24, assemble 24–36. Transfers of three: cut 0–12, machine 3–15, assemble 6–18 — the second step starts on the first three while the first is still cutting the rest. Nothing runs faster; the steps overlap.

EPE: Every Product Every ___

EPE (Every Product Every interval) is the lean metric for batch size flexibility. It measures how often you can cycle through your full product mix:

EPEMeaningTypical Changeover Capability
EPE MonthEach product runs once per monthLong changeovers (hours). Large batches required.
EPE WeekEach product runs once per weekModerate changeovers. Typical starting point.
EPE DayEach product runs every dayShort changeovers (<15 min). True leveling possible.
EPE ShiftEach product runs every shiftVery short changeovers (<5 min). Near one-piece flow.

The Goal: EPE Day (or Better)

When you can make every product every day, lead time drops to 1 day, finished goods inventory drops to 1 day of supply, and you can respond to any customer demand within 24 hours. EPE Day is the practical target for most lean operations. It requires changeovers under 10-15 minutes — which is exactly what SMED delivers.

Calculating Optimal Batch Size

The traditional Economic Order Quantity (EOQ) formula optimizes for setup cost vs. holding cost. But in lean thinking, the approach is different: reduce setup time, then reduce batch size.

Reduce Changeover
(SMED)
enables →
Smaller Batches
creates →
Less WIP +
Shorter Lead Time
enables →
Better Leveling +
More Flexibility
SMED is the enabler. Batch reduction is the result. Flow, flexibility, and short lead times are the reward.

Use the batch size calculator to find your current optimal batch given your changeover time, then see how reducing changeover time enables smaller batches.

The Lead Time & Batch Size Connection

Lead time is directly proportional to batch size. Halve the batch, halve the lead time. The math is simple:

Lead Time = Batch Size × Cycle Time + Queue Time

A batch of 500 at 30 seconds/unit = 4.2 hours of processing + queue time at every step. A batch of 50 = 25 minutes of processing + much shorter queues (because smaller batches flow faster through the system). Total lead time drops 80-90% when batch size drops 90%.

Reducing Batch Size Step by Step

Measure your current EPEHow often does each product run today? If Product A runs once every 3 weeks, your EPE is 3 weeks. That is your baseline.
Apply SMED to the longest changeoverReduce the changeover that is gating your batch size. Target: cut it in half as a first step. See SMED guide.
Cut batch size by 50%With faster changeovers, reduce batch size by half. Track the impact: WIP should drop, lead time should shrink, and schedule flexibility should improve.
Repeat until EPE DaySMED again, reduce batch again. Each cycle gets easier because smaller batches expose problems faster and the team learns to changeover more efficiently.
Implement heijunkaOnce EPE is 1 day, you can level-load the schedule — every product every day in a repeating pattern. This is the lean scheduling ideal.
✅ Small Batch Thinking
  • Batch size driven by demand, not setup avoidance
  • Changeovers reduced to enable flexibility
  • EPE tracked and improving over time
  • Lead time and WIP declining
  • Can respond to demand changes within 1-2 days
❌ Large Batch Thinking
  • "We cannot change over more often" (have you tried SMED?)
  • Batch size set to minimize changeovers per week
  • 3-6 week EPE — customers wait weeks
  • Warehouse full of finished goods "just in case"
  • Schedule changes require weeks of notice

🎯 Key Takeaway

Batch size is the lever that connects changeover time to lead time, WIP, and flexibility. Reduce changeovers with SMED, then reduce batch sizes. Each halving of batch size halves your lead time and WIP. Target EPE Day — every product every day — and unlock the ability to level-load your schedule, respond to customer demand in 24 hours, and run a calm, predictable operation.

Interactive Demo

Adjust batch size to see how setup costs, holding costs, and lead time change. Find the optimal economic order quantity.

⚡
Try It Yourself
Batch Size & EOQ Explorer
▼
Adjust batch size to see the trade-off between setup costs (fewer batches = less setup) and holding costs (larger batches = more inventory). The EOQ minimizes total cost.
40 units
5 units100 units
1000 units
200 units3000 units
$200
50$500$
$5/yr
1$/yr20$/yr
$0$10003$20006$30009$4001310255075100Batch Size (units)EOQ=283
Setup Cost
Holding Cost
Total Cost
EOQ Point
Lead Time Impact (proportional to batch size)
ShortLong
$5100$
▲ 3686$ vs baseline
Total Cost
$5000$
Setup Cost
$100$
Holding Cost
283 units
EOQ
Ready for the full knowledge check? Test your understanding with guided scenarios and data export.
PROTake the Pro Knowledge Check →
Free forever · Every feature included

Stop reading, start modeling

Model your process flow, run simulations, optimize staffing with TOC math, and test your knowledge with 107 interactive checks — all in one platform.

Open Workbench → Pro Knowledge Check

Take this to a room

The running order

For planners and supervisors who set run quantities. They should leave able to state their EPE and name the changeover holding it there.

6 beats · 10 min
  1. 1

    How we pick a number today

    Ask how the run quantity gets decided. Almost everyone answers with setup time, and that is the whole point.

    • Setup cost is the only batch cost that appears on the sheet.
    • Lead time, WIP, space and late scrap are real, and none of them are on it.

    Ask the room What was the biggest batch we ran last week, and who chose the number?

  2. 2

    The costs that are not on the sheet

    Six costs ride along with a big batch. The setup calculation cannot see any of them.

    • A batch of 1,000 takes ten times longer to finish than a batch of 100.
    • The pile between two steps is the batch size — that is cash and floor space.
    • A defect that starts at unit 50 is found when unit 1,000 is done: 950 suspect.
  3. 3

    Lead time is proportional. Full stop.

    Halve the batch and you halve the lead time. This is arithmetic, not an improvement programme.

    • 500 units at 30 seconds is 4.2 hours of processing before anything ships.
    • 50 units is 25 minutes — and the queues behind it shrink as well.
    • Total lead time typically falls 80–90% when batch size falls 90%.

    Ask the room If we halved our biggest batch tomorrow, what would actually break?

  4. 4

    EPE: the number to argue about instead

    Stop debating batch size and measure how often each product runs. That is EPE — every product every what.

    • EPE month: changeovers run into hours, so batches must be large.
    • EPE week: the usual starting point for a plant that has never measured it.
    • EPE day: changeovers under 15 minutes, and levelling becomes possible.

    Ask the room What is our EPE right now — how many days between two runs of our top product?

  5. 5

    The gate is changeover, not batch size

    You cannot cut the batch until you cut the setup. SMED comes first and the batch follows it down.

    • Shorter changeover → smaller batch → less WIP and shorter lead time → levelling.
    • Batch reduction is the result. Changeover time is the lever.
    • Cutting the batch without cutting the setup just buys lost capacity.
  6. 6

    What we do on Monday

    Four steps, and the first one takes an afternoon.

    • Baseline the EPE for the top five products this week.
    • Pick the changeover gating the biggest batch and halve it.
    • Halve that batch, then watch WIP and lead time for two weeks.
    • Repeat the pair until EPE is one day, then level the schedule.

    Ask the room Which changeover do we attack first, and who owns it by Friday?