Skip to content
allanninal.dev/build
Collaborate

Part 4 of 7 · Stock transfer planner series ~5 min read

How transfers get batched onto movements that already happen

The single change that makes stock transfer worthwhile in most businesses is not better detection; it is waiting until Thursday.

Key takeaways

  • A proposal waits for a scheduled movement rather than triggering one.
  • Set a maximum wait; beyond it the benefit has decayed and the proposal expires.
  • Assemble the whole load for a movement at once, so picking is one trip round the racking.
  • A dedicated trip needs a benefit several times larger, and sometimes that exists.
  • Urgent transfers bypass all of this, and should be counted.

Waiting for a van

How a transfer proposal is attached to a scheduled vehicle movementA vertical chain of five steps entered by a box labelled A transfer clears on the economics. Step one asks whether a movement from A to B is scheduled in the next ten days; if not it exits to Hold, rechecking daily. Step two asks whether there is space on it by volume and weight; if not it exits to Next one, or split it. Step three asks whether it is still needed, since B may have sold out or been replenished; if not it exits to Cancel, quietly. Step four adds it to the load with everything else going. Step five produces one pick list for the whole movement. A note says the third gate matters, because a proposal made nine days ago may have been overtaken.AWS ACCOUNTA transfer clearson the economicsMovement scheduled?A to B, next 10 daysHoldrecheck dailynoSpace on it?volume and weightNext oneor split itnoStill needed?B may have sold outor been replenishedCancelquietlynoAdd to the loadwith everything else goingOne pick listfor the whole movementThe third gate matters: a proposal made nine days ago may have been overtaken.
Fig 1. How a proposal becomes a picked transfer. Rechecking at the point of picking prevents the most common form of wasted movement.
  • Compute
  • App integration
  • Machine learning
  • Security & identity
  • Management
  • Front-end & mobile

One pick list per movement

This is where a second saving hides. Twelve separate transfers picked individually is twelve trips round the warehouse; the same twelve as one consolidated pick list, ordered by location, is one trip and perhaps a third of the labour.

Which changes the economics again, and in the right direction: the per-transfer picking cost falls as the batch grows, so a set of marginal transfers can become collectively worthwhile when none of them would be individually. The economic test therefore runs against the batch as well as the item.

How long to wait

How the benefit and cost of a transfer change with waiting timeA stacked bar chart with four bars in pounds. Two series: benefit remaining in green, and cost of the transfer in red. Waiting zero days: twenty-eight pounds of benefit against forty-two pounds of cost. Waiting four days: twenty-four against eleven. Waiting eleven days: fourteen against eleven. Waiting twenty-five days: four against eleven. A note says waiting saves transport and spends benefit, and four days is the sweet spot here.£0£25£50£75£100~£70Wait 0 days~£35Wait 4 days~£25Wait 11 days~£15Wait 25 daysBenefit remaining, £Cost of the transfer, £Waiting saves transport and spends benefit. Four days is the sweet spot here.
Fig 2. The trade between waiting and acting. Benefit decays because the destination either loses the sales or gets replenished normally, so an indefinite hold is not free.

The decay is real and has two causes. The destination may lose the sales it was going to protect, and it may get replenished from the supplier in the normal course, at which point the transfer is unnecessary.

So proposals expire. A maximum hold of around the destination’s own replenishment lead time is a defensible default: beyond that, the normal supply chain will have solved it and the transfer is redundant.

When a dedicated trip is justified

Occasionally the benefit is large enough: a high-value item, a customer waiting, a seasonal peak where the sales genuinely will not happen otherwise. Those exist and the system should be able to say so, with the arithmetic showing why this one clears a much higher bar.

The bar should be several times the transfer cost rather than merely above it, because the estimates on the benefit side are soft and a dedicated trip commits real money against a soft number.

Urgent transfers

The ones that bypass everything

  • A customer is waiting and the sale is confirmed. Move it.
  • A production line is stopped. Move it.
  • These do not need an economic test because the benefit is known and large.
  • They do need recording as urgent, separately from planned transfers.
  • Because the count is a signal: a lot of urgent transfers means the planned ones are not happening, or the stock allocation is wrong.
  • Never make the urgent path harder than picking up the phone. It will simply be routed around.

The counting point is the useful one. Urgent transfers are expensive — they are dedicated trips by definition — and a business running twenty a month is spending significantly on something that a better allocation would largely remove.

Next: where the stock actually is while it is moving.

All posts