VMI and consignment answer two different questions

The actual difference between the two terms

The difference is when ownership transfers, and consequently who carries dead stock.

Why fasteners in particular reward this kind of arrangement — the cost structure behind it: the 15-85 rule.

Four combinations, not two

Separate the two questions and there are four possible arrangements:

The bottom-right cell is where the negotiation happens: the supplier carries both the capital and the replenishment decision. It is the best outcome for the customer, and suppliers do not give it away — it is usually traded for a longer contract term, firmer volume forecasts, or compensation in the unit price.

It succeeds or fails on the consumption data

Both models rest on the same premise: your consumption data is trustworthy. The supplier decides how much to replenish using your usage figures. If the data is not clean, the results are predictable:

• Over-replenishment — the supplier raises safety stock to avoid a stockout, and the shelves fill with low-value parts
• Under-replenishment — usage is underestimated, shortages still happen, and now responsibility is blurred
• Counts that do not reconcile — especially awkward under consignment, because it directly determines who owes what

So there is one question to answer honestly before starting: are your issue records real-time, or entered at month end?

If it is the latter, this can still be done — but the consumption reporting method has to be agreed first. Otherwise the supplier is simply making decisions on your bad data.

Six things the contract has to settle

These six belong in writing before the first stock arrives. Negotiating them afterwards is usually an argument.

1. The point at which ownership transfers. On delivery, on consumption, or deemed sold after sitting for some period? The third is a common compromise — the supplier carries the risk, but with a time limit.
2. How consumption is recognised, and how often it is reported. Barcode, weight, or manual reporting? Pulling directly from the work order system is ideal. This item determines the accuracy of the whole mechanism.
3. Who sets the stock levels, and how they change. What are the initial min/max based on? How often are they reviewed as usage shifts, and who has authority to change them?
4. Liability and compensation for a stockout. If a line stops because replenishment was late, how is that established and settled? Without this clause, VMI is just handing the management nuisance to someone else with no substantive protection.
5. Dead stock and specification changes. When a design revision retires a part number, what happens to the stock on the shelf? In industries that revise frequently this is certain to happen, not hypothetical.
6. The exit. On termination or expiry, who buys the remaining stock, within what period is it cleared, and how is the count conducted. The arrangements that work are usually the ones where the break-up was negotiated too.

Where to start a trial

Not everything at once. A sensible start is a group of part numbers meeting three conditions:

• Stable usage — items with volatile demand make a poor first batch and cost both sides confidence in the model
• Enough line items — the management saving only becomes visible at some breadth; three part numbers will feel like nothing
• No imminent revision — avoid hitting a dead stock dispute during the trial itself

This is not one of the six steps. It shows up across them, or after assembly. Where the decisions that lead here were made is in specifying a screw, which sets out the order and why doing it out of order is rework.

Common questions

Is VMI cheaper?

The unit price usually is not cheaper and may be slightly higher, because the supplier is taking on the replenishment decision, the inventory management and sometimes the capital cost, and those appear in the quotation. What falls is indirect cost: order count, goods-in inspection hours, stock counting labour, dead stock and expedited shipments. So an evaluation cannot compare unit prices alone, or it will reach the wrong conclusion that VMI costs more.

What does a supplier gain from consignment, and why would they agree?

Consignment does tie up the supplier’s capital and carry their risk, but it buys stickiness and visibility. Stock sitting at the customer occupies the position ahead of competitors, and the supplier sees real consumption data, which makes their own material planning and scheduling considerably more accurate. So it is normally an exchange — contract term, volume commitment, exclusivity or priority of supply. If you are asking for consignment, it helps to arrive with something you can offer.

Our ERP has no real-time issue data. Can we still do this?

Yes, but the reporting method has to be settled first rather than pretending the problem does not exist. A common transitional approach is periodic manual count reporting, weekly for example, with min and max levels set conservatively, tightening them once a few months of data have accumulated. What matters is that both sides acknowledge the data limitation at the outset and write the reporting frequency into the contract. The worst approach is claiming the data exists when it is actually entered at month end, because the supplier then makes wrong decisions continuously.

We revise specifications frequently. Is this still suitable?

Yes, but the sixth item — dead stock and specification changes — becomes the most important clause and cannot be left vague. The practical approach is to set a notice period for revisions and a method for apportioning the stock on the shelf when one happens, for example stock within a few weeks of notice taken by the customer and the remainder handled by the supplier. Frequent revision is not itself an obstacle; failing to write revision into the contract is.

References

• Inventory Operations Consulting — vendor managed inventory: how it operates and its common problems (independent inventory management reference)

Acceptance and commercial terms for any particular arrangement are governed by your own contract.