The Reorder Point That Belongs to No Warehouse
Reorder point, safety stock and lead time all live on the item, and nowhere else. Every low-stock check in the product compares the national total against them. A distribution centre can be at zero for a fortnight while the number on the dashboard stays green.
The short version
One reorder point per item, compared against total stock across every warehouse you own. If you run one store, this is correct and you can stop reading. If you run regional distribution centres two thousand kilometres apart, the low-stock alarm is measuring the wrong thing and you need a per-site discipline that the system will not enforce for you.
A distributor with centres in São Paulo, Recife and Manaus does not have one stock position. It has three, and the distance between them means they are, for practical purposes, three businesses that share a purchasing department.
Our system knows exactly how much is in each. It holds stock per warehouse and per bin location, moves it with a full history, and will report the balance at any site you ask about. What it will not do is worry about any of those numbers individually.
One number, one alarm
An item carries a reorder point. It also carries a safety stock figure and a lead time, with organisation-wide defaults behind them. Those three are the inputs to every replenishment signal in the product.
And every one of those signals — the low-stock report, the dashboard tile, the reorder list, the count of items needing attention — is the same comparison: total stock across all warehouses, against the item's single reorder point. There is no per-warehouse variant of that comparison anywhere, because there is no per-warehouse reorder point for it to use.
The system can tell you precisely where your stock is. It cannot be told where you need it.
The failure has a shape, and it is not random
It always fails the same way: the site that is short is invisible while the network is long. Which is the common case, because stock accumulates where it lands and demand happens where the customers are.
One item, three centres, one reorder point
Nothing here is a defect in the arithmetic. 460 is genuinely more than 300. The problem is that 460 is not a quantity anybody can sell, because the units are in the wrong cities and the freight between them takes days.
Now run it in the other direction and it is just as bad. Set the reorder point high enough that São Paulo never runs dry — say 900 — and every one of your items is permanently in alarm, on every screen, for the whole network. An alarm that is always on is an alarm that gets turned off.
Why the distances make this a Brazilian problem specifically
Not because the software behaves differently here, but because the cost of being wrong scales with the distance between your sites, and few distribution networks in the world are stretched further.
Rebalancing between two warehouses in a small country is an inconvenience — a van, an afternoon. Rebalancing between the south-east and the north is a decision with a freight cost, a transit time measured in days rather than hours, and in some corridors a river. A system that lets a site drift to zero without saying anything is not costing you a phone call; it is costing you the difference between replenishing from your own network in time and buying locally in a hurry.
The discipline that works instead
-
Set the reorder point for the site that matters most, not for the network
Pick your largest-demand centre and set the item's reorder point at what that site needs. You will over-order slightly for the network and you will stop running the one site dry. This is the single highest-value change on this page and it costs nothing.
-
Run a per-warehouse stock report on a fixed day, every week
The data is there and the report will give it to you per site. What is missing is anything that pushes it at you, so make it a standing item with a named owner rather than something somebody remembers.
-
Treat rebalancing as a scheduled activity, not an exception
Transfers between your own warehouses are fully built — dispatch, receipt, partial receipt, shortfall, return to source. What is not built is anything that suggests a transfer. Put a weekly rebalancing review in the calendar and use the per-warehouse report as its agenda.
-
Do not let the dashboard tile be your replenishment process
It is measuring the network. If your business is regional, it is answering a question you are not asking, and it is answering it with a green tick.
Four questions about replenishment, for us or anybody else
Show me the reorder point for this item at this warehouse.
A good answer sounds like
A field, on the warehouse-item combination, with its own value.
What it actually means
If the only reorder point is on the item, the alarm is national. Ours is. Ask early rather than discovering it from a stockout.
Which stock figure does the low-stock alert compare against?
A good answer sounds like
A named figure — total, on-hand at a site, or available after allocations.
What it actually means
Three different answers, three different products. The word "available" is doing a lot of work in most demonstrations.
Does anything suggest a transfer between my own sites?
A good answer sounds like
A rebalancing suggestion, or an honest no.
What it actually means
Suggestion engines are rarer than vendors imply. What matters is whether the per-site report exists to do it manually.
Can lead time differ by supplier and by destination?
A good answer sounds like
Yes, held on the supply relationship.
What it actually means
Ours is one figure on the item, with an organisation-wide default behind it — so a two-day delivery and a six-week import share a number.
Per-site replenishment is a dimension, not a new module
The stock is already held per warehouse and per location. What is missing is a place to put a target against that same dimension, and consumers that read it. That is a well-shaped piece of work rather than an open-ended one, and it is worth describing precisely so you can judge the size of it.
A reorder point per warehouse-item
One row, one target, and the low-stock comparison run against the per-warehouse balance the system already maintains.
A rebalancing suggestion
Where one site is below its target and another is above, with the transfer machinery that already exists on the other end of it.
Lead time on the supply route
The smallest of the three and the one that quietly improves every other number, because a single item-level lead time cannot describe a network with an import and a local supplier in it.
No dates on a public page. Describe your network — how many sites, how far apart, and which one hurts when it is empty — and we will come back with a written scope, timeline and cost before you commit.
Scope per-site replenishmentWhat AWRA OpsHub does today
- Stock held, valued and reported per warehouse and per bin location, with complete movement history.
- A reorder point, a safety stock figure and a lead time on every item, with organisation-wide defaults behind the last two.
- Low-stock reporting, a dashboard tile and a reorder list, all driven from the item's own figures.
- Automatic requisition raising against a department when an item falls below its point, through the workflow engine.
- Transfers between warehouses with dispatch, receipt, partial receipt, recorded shortfall and return to source.
- A blended 30/90-day consumption velocity per item with a suggested reorder quantity built from lead time and safety stock.
What it does not do
- Any reorder point, safety stock or lead time held per warehouse or per location. All three live on the item and nowhere else.
- A low-stock comparison against a per-site balance. Every check in the product compares the total across all warehouses.
- Rebalancing suggestions between your own sites.
- Supplier selection on an automatic replenishment — the rule raises a requisition against a department and sourcing stays manual.
- Order sizing beyond a fixed configured quantity or the difference between the reorder point and current stock. No pack sizes, no price breaks, no demand shaping.
Not ours, by choice
- One reorder point is the right design for a single-site business, and most businesses are single-site. This page is about the ones that are not.
- Setting the point for your largest site is a real mitigation and we recommend it, but it does over-order for the network and you should know that you are choosing it.
- Nothing here is Brazilian. It is the arithmetic of a spread-out network; Brazil is where the distances make it expensive fastest.
Size the reorder points with somebody
The hour that pays for itself is the one where you decide which site each item's reorder point is really protecting. Bring your three biggest centres and your top hundred lines.
Set the points properlyFrequently asked questions
Can I use a custom field to hold a per-warehouse reorder point?
You could store the number, but nothing would read it. Custom fields capture information; they do not change what the low-stock comparison compares. The alert would still be measuring the national total.
Does the automatic requisition know which warehouse is short?
It fires on the item-level comparison, so it knows the item is below its point across the organisation. It raises a requisition against a department, and it does not select a supplier or a destination site — sourcing and routing stay with your buyer.
Is a per-warehouse balance report available today?
Yes. Stock is held per warehouse and per location and can be reported that way, over any period. The gap is not visibility, it is that nothing compares those figures to a target or pushes them at you.
What should I set the reorder point to in the meantime?
To the requirement of the site whose stockout costs you most, not to the sum of the network. You will carry slightly more inventory in total and you will stop discovering an empty distribution centre from a customer.
