Kanban and Min/Max for MRO Stockrooms
Reorder while usable stock still covers the time needed to refill it. Two-bin Kanban uses an empty working bin as the signal. Min/max uses a counted reorder threshold. Both need a named owner, known response time, and visibility of orders already in progress.
In this guide
Two bins: empty the working bin, then use the reserve
The empty bin starts the order. The reserve supports work while that order is completed.
- 01 / USEWorking bin
Take stock from this bin first.
- 02 / SIGNALWorking bin empty
Send the card. Mark refill pending.
- 03 / COVERReserve in use
Cover the full time until refill.
- 04 / CLOSEStock received
Check, put away, restore the reserve.
Kanban is a visible replenishment signal, such as a card or empty container. In a two-bin arrangement, keep working stock and reserve stock distinct. Do not dip into both without recording it; the reserve then stops representing the coverage you planned.
1. Take parts from the working bin.
2. When it empties, send its card or signal to the refill owner. Record that the refill is pending.
3. Use the reserve while replacement stock is ordered, delivered, checked and put away.
4. Refill to the agreed quantity, restore the working/reserve arrangement, and close the signal.
The bins do not have to hold equal quantities. The reserve must cover the full response window. If both bins are empty, the system has already missed its intended signal.
Include the time before anyone places the order
Reserve coverage = expected use during the full response window + a chosen buffer.
Suppose a shop uses 8 pieces per week. A weekly review can delay action by up to 1 week. Ordering through receipt takes another 2 weeks. The chosen buffer is 8 pieces.
| Step | Calculation | Result |
|---|---|---|
| Full response window | 1 week review delay + 2 weeks order-to-available time | 3 weeks |
| Expected use in that window | 8 pieces/week × 3 weeks | 24 pieces |
| Planning reserve | 24 + 8 buffer | 32 pieces |
The 8-piece buffer is an explicit assumption, not a universal safety-stock rule. Select it from variation, acceptable shortage risk, delivery reliability, and consequence. A planned shutdown pull should be handled as known demand instead of being hidden in the average.
Keep the units consistent. Pieces per week must be multiplied by weeks, not by a lead-time number written in days. Review the plan after shortages, unusual pulls, or supplier changes. For intermittent critical spares, average usage alone may be misleading.
Min/max: count what is available and what is already coming
Before ordering, calculate inventory position: usable on hand + confirmed incoming stock − allocated demand.
For this example policy, reorder when inventory position is at or below the minimum; order toward the maximum. Other systems may use different trigger rules, so write down which rule your shop follows.
| Example item | Value |
|---|---|
| Minimum trigger | 32 pieces |
| Maximum target | 64 pieces |
| Usable on hand | 26 pieces |
| Confirmed incoming order | 20 pieces |
| Already allocated to a job | 6 pieces |
| Inventory position | 26 + 20 − 6 = 40 pieces |
Position is 40, above the 32-piece trigger. Ordering another refill merely because the bin contains 26 pieces would duplicate coverage. This arithmetic assumes the incoming 20 arrive when needed; a late order still requires action.
If there were no incoming order, position would be 26 − 6 = 20. The unrounded order would be 64 − 20 = 44 pieces. A 10-piece supplier pack would round that to 50, making planned position 70. That exceeds the nominal target by 6; check approved limits and physical capacity before accepting the rounding. Do not quietly call 70 a 64-piece maximum.
Put a status and an owner on the signal
A card moved to a desk is not proof that an order was placed.
Use a short visible status chain: signal received, order placed, due date confirmed, received, put away. Record the exact item and quantity at each handoff. Assign a backup for absence and a route for overdue orders.
A useful card carries the item specification, location, refill quantity or rule, supplier/internal item number, and responsible person. Keep pending cards visible until the right material is usable in the right location. Separate the arrival of a box from completion of the refill.
Count frequency must fit demand and coverage. A monthly check is inadequate if the remaining stock can disappear in a week. A vendor-managed route needs agreed service scope, visit timing, ownership and escalation; the label should keep the same item identity whichever person performs the refill.
Know which items need more than a simple card
Keep controlled parts, shelf-life limits, and one-time projects visible as exceptions.
Do not automatically reorder an obsolete OEM component, expired adhesive, unapproved PPE substitute, or traceable assembly from a generic bin signal. The signal can start a review without authorizing the purchase.
Investigate repeat shortages: missed signals, unrecorded borrowing, large job pulls, supplier delays, inaccurate counts, or reserve stock used early. Raising every maximum may conceal the cause while filling the wrong shelves.
Frequently Asked Questions
Is the minimum the amount we should always have left?
It is the trigger under the chosen policy. Stock may fall below it while a refill is in progress. The coverage calculation must account for that time.
Can we use one bin?
Yes, with a reliable count or level signal. Two bins make the reserve physically visible; they do not remove the need for correct quantities and an owner.
Does an incoming order prevent a shortage?
Only if the correct usable stock arrives before demand consumes what remains. Check its due date, not just its existence.
Sources and Review Boundaries
Published by GO4FORGE. Revised 2026-10-29. Editorial method. These notes help narrow sourcing evidence and selection risk; they do not replace the responsible site authority, OEM procedure, code requirement, or qualified engineering/safety review.
- GO4FORGE product knowledge files Internal product-knowledge references distilled into public decision guides.
- GO4FORGE RFQ evidence model Shop-floor evidence patterns for quotes, substitutions, and sourcing review.
- GO4FORGE hardware replenishment knowledgebase Two-bin operating sequence and refill ownership. All numerical examples are hypothetical planning examples, not measured demand or service guarantees.
Related Guides
Keep recurring parts within reach
A bin keeps running out? Tell us where it is and what goes in it. Ask FORGE about this topic.