Can consumption signals and planning coexist for the same part?

They can, and it is a normal arrangement: the planning system holds the item for costing, traceability and long-horizon visibility, while the physical refill on the floor runs off a signal rather than a released order. What must be avoided is two mechanisms independently ordering the same part, which is how plants end up with unexplained excess. Configure the item so that floor replenishment consumes against the plan rather than adding to it, and make sure receipts post once.

Does a card system require a formal improvement programme first?

It requires the conditions the loop assumes, which is a narrower thing. Consumption has to be reasonably repetitive, the supplying step has to be dependable within the replenishment period, and the physical discipline of returning signals and using the right containers has to hold across shifts. Those conditions are frequently created through improvement work, but they can also exist naturally in a stable operation. Where they do not exist, installing loops produces stoppages that get blamed on the method rather than on the instability.

What is the honest weakness of planned replenishment on a shop floor?

Distance from reality. The calculation is only as good as the parameters and balances behind it, and shop floor conditions move faster than master data gets updated. The symptom is familiar: a plan that says material is available while the line is short, and a shortage meeting that has become a permanent institution. Planning excels at forward commitment and aggregation across products, and it is weakest at reflecting what is physically on a rack at this moment.

Refilling by signal or by calculation on the shop floor

Two mechanisms for the same job: getting the right part to the right station before the line needs it. One reacts to what has physically been used, using a signal that travels back up the flow. The other calculates what will be needed from a schedule and a bill of material, and issues orders in advance. Their data requirements are almost opposites, which is why plants tend to be good at one and quietly bad at the other.

Comparison criteria

Criteria are stated explicitly and neither option is declared a winner: which one fits depends on the constraint that binds hardest in your operation.

Criterion	Kanban: replenishment triggered by consumption	MRP: replenishment calculated from schedule and bills of material
What triggers a refill	A physical or electronic signal generated the moment stock is drawn: an emptied bin, a returned card, a scanned pick at the point of use.	A planned order created by netting requirements against on-hand and on-order balances, offset backwards by the item's stated lead time.
Master data the method rests on	Very little in the system. Accuracy lives in the physical arrangement: container quantity, number of loops, and the discipline of returning signals.	A great deal. Bill of material accuracy, routings, lead time parameters, order policies and inventory balances all have to reflect reality.
Behaviour with steady demand	Excellent. Repeated consumption is precisely what the loop was sized for, and the mechanism self-corrects as usage rate drifts slowly.	Works, but with a planning overhead that adds nothing for an item consumed at a constant rate day after day.
Behaviour with lumpy or one-off demand	Poor without intervention. A loop sized for routine use cannot absorb a step change, and nothing in the mechanism anticipates the future.	Strong. A future requirement enters the schedule and is committed ahead of consumption, which is exactly what a calculation can do and a signal cannot.
Long lead time and imported items	Awkward. Covering a long replenishment period means many containers in circulation, which is inventory carried permanently to hold a signal.	Its natural territory. Requirements can be committed months before consumption, which is the only way a distant supplier can be given notice.
What an engineering change does to it	Leaves obsolete stock sitting in loops that keep replenishing themselves until somebody physically withdraws the cards and empties the bins.	Handles phase-in and phase-out deliberately through effectivity in the bill of material, provided the change is entered correctly and on time.
Visibility and audit trail	Immediate on the floor and thin in the record. Anyone can see the state of a loop; reconstructing what happened last month is harder.	Recorded end to end, with planned orders, exception messages and receipts available for analysis, costing and traceability.
How each one goes wrong	Loops sized once and never revisited, cards lost, containers used for something else, and consumption rates that quietly moved away from the sizing.	Parameters nobody maintains: an optimistic lead time, a stale order quantity, an inventory balance that disagrees with the shelf, producing plans that look precise and are not.

Choose Kanban: replenishment triggered by consumption when

The item is consumed repeatedly and at a broadly steady rate across the working week
The supplying cell or vendor can refill within the loop's replenishment period
Consumption happens at a physical point of use where a container can hold the signal
System inventory balances are unreliable and the shelf is the more honest record

Choose MRP: replenishment calculated from schedule and bills of material when

Procurement lead times are long enough that commitment must precede consumption
Demand is intermittent, project-driven or specific to an engineered order
A change has to be phased in across many items on a controlled effectivity
Components are shared across many end products and requirements must be aggregated

Sizing the loop is the entire engineering content of a card system

A signal-based refill works only if the quantity in circulation covers consumption across the replenishment period, plus enough margin for the variation the plant actually experiences. Get it small and the line stops; get it large and you have built a warehouse in the aisle. The sizing depends on usage rate, replenishment time, container size and the reliability of the supplying step, and every one of those moves over a product's life. Plants that treat sizing as a one-time calculation at implementation end up with loops that are wrong in both directions, and the visible failures push everyone back to expediting. Schedule a review of the sizing on a fixed cadence and after any change to volume or supplier.

Planned orders inherit the quality of parameters nobody owns

A planning run does exactly what its inputs tell it to. If a lead time in the item master reflects what a supplier promised years ago rather than what they deliver now, every order gets released too late and the shortage list becomes the real plan. The same applies to order quantities set once during implementation, scrap factors nobody revisits, and inventory balances that drift away from what is on the shelf. None of these show up as an error message; they show up as planners overriding the system and eventually distrusting it. Assign explicit ownership for lead times, order policies and record accuracy, and audit a sample of them regularly rather than waiting for a shortage.

The usual design uses both, split by lead time and demand pattern

A common and effective arrangement plans purchased items with long or uncertain supply through the system, where forward commitment is possible, and refills high-usage local and internally made items by consumption signal at the point of use. Fasteners, packaging and standard components sit naturally in loops; castings from overseas, custom electronics and anything engineered do not. The boundary is not permanent. When a supplier is brought closer, an item can move to a loop; when demand for a family becomes erratic, it should move back into the plan. What matters is that each item has a deliberate assignment, rather than an accident of which system was implemented most recently.

Frequently asked questions

Can consumption signals and planning coexist for the same part?: They can, and it is a normal arrangement: the planning system holds the item for costing, traceability and long-horizon visibility, while the physical refill on the floor runs off a signal rather than a released order. What must be avoided is two mechanisms independently ordering the same part, which is how plants end up with unexplained excess. Configure the item so that floor replenishment consumes against the plan rather than adding to it, and make sure receipts post once.
Does a card system require a formal improvement programme first?: It requires the conditions the loop assumes, which is a narrower thing. Consumption has to be reasonably repetitive, the supplying step has to be dependable within the replenishment period, and the physical discipline of returning signals and using the right containers has to hold across shifts. Those conditions are frequently created through improvement work, but they can also exist naturally in a stable operation. Where they do not exist, installing loops produces stoppages that get blamed on the method rather than on the instability.
What is the honest weakness of planned replenishment on a shop floor?: Distance from reality. The calculation is only as good as the parameters and balances behind it, and shop floor conditions move faster than master data gets updated. The symptom is familiar: a plan that says material is available while the line is short, and a shortage meeting that has become a permanent institution. Planning excels at forward commitment and aggregation across products, and it is weakest at reflecting what is physically on a rack at this moment.

Data limitations

Manufacturing figures are operator-supplied inputs, not market data. GeoBusinessIQ holds no factory costs, production volumes, yields, cycle times, tooling prices or capacity data and does not estimate them — every result reflects only the figures you enter.
No manufacturer, supplier, vendor or factory is recommended, rated or ranked anywhere in this cluster, and no directory of them is published. Selection material describes how to run your own assessment; the assessment itself remains yours.

Explore the graph

Logistics & supply chain

Across GeoBusinessIQ

Sources

NIST Manufacturing Extension Partnership — NIST MEP (accessed 2026-08-30)
Covers: A public programme supporting small and medium manufacturers with operational, quality and technology adoption practice.
Does not cover: Results attributable to any specific manufacturer, or improvement figures transferable to another plant.
Why it matters: Cited for the operational practice it publishes for smaller manufacturers, not for benchmarks or outcome claims.
Review cadence: annual
United Nations Industrial Development Organization — UNIDO (accessed 2026-08-30)
Covers: Industrial development analysis, industrial statistics methodology, and manufacturing capability programmes across member states.
Does not cover: Company-level data, factory costs, supplier information, or real-time production statistics.
Why it matters: The United Nations agency for industrial development; used for structural framing of how manufacturing sectors develop, never for point figures.
Review cadence: annual
OECD — OECD — economic and tax statistics (accessed 2026-05-14; reviewed 2026-05-18)
Covers: Comparable corporate tax, statutory rate, and economic indicators across member and partner economies.
Does not cover: Effective tax rates, deductions and incentives, local surtaxes, and personal residency rules.
Why it matters: Used as a cross-country baseline to sanity-check rates against primary tax-authority figures.
Review cadence: Annual, plus on major statutory changes.

Educational and operational information only — not legal, engineering, safety, customs, tax, or financial advice. Requirements vary by jurisdiction, product, process, and contract; confirm with the relevant authority or a qualified professional before acting.

Last updated: 2026-08-30