When to Reorder and How Much: Reorder Point and Safety Stock

When to Reorder and How Much: Reorder Point and Safety Stock

The first three parts built the foundation. We looked at why the recorded balance never matches the shelf, then closed that gap by capturing every movement as it happens through scanning, then tied consumption to a production order with an actual cost. The result is not prettier reports. It is something far more practical: you now have a trustworthy consumption history for every item.

That history is what turns the two hardest decisions in a warehouse from guesswork into arithmetic. When do I order? And how much? Today, in most factories, both questions are answered by a single sentence: when the storekeeper says an item is nearly finished. That sentence costs you on both sides at once, because it produces stockouts on critical items and dead stock on others at the same time.

Why ordering by instinct fails specifically

Ordering by instinct is not random. It draws on real experience. Its problem is that it depends on one person, treats every item alike, and has no way to handle variability. So it shows up in three recurring forms:

• Ordering too late: someone notices the item when it is nearly gone, so every purchase becomes an emergency: a higher price, faster and costlier freight, and sometimes an alternate supplier of different quality.
• Ordering too much: the item is bought in large quantities to guard against a stockout or to chase a volume discount, and capital turns into goods sleeping on a rack, occupying space another item actually needs.
• Treating every item the same: the item that shuts down a production line is managed exactly like the item you can buy locally within the hour. Attention is spread evenly while the consequences are not.

The alternative is not a system that orders on your behalf. It is three numbers per item that you calculate once and review on a schedule.

The first number: average consumption

How many units of this item do you consume per day or per week? This number does not come from memory. It comes from the issue history you built in part two. Two things matter here.

First, choose a period that represents your reality. An average calculated over a single month containing a planned shutdown or a peak season gives you a misleading figure. Second, watch the trend, not just the average. An item whose average has held steady for a year is a different case from one whose consumption climbs every month because a product is selling more. The average tells you where you stand; the trend tells you when the number is due for review.

The second number: lead time and its variability

Lead time is the elapsed time between raising a purchase order and the goods being available on the shelf, ready to issue. This is where the most expensive mistake happens: factory owners calculate lead time as the supplier's promise alone, while the reality also includes internal approval of the request, freight, customs clearance for imported items, and inspection and receipt at the store.

More important than the average lead time is its variability. A supplier who always delivers in two weeks is far better than one who sometimes delivers in one week and sometimes in five, even when the average is identical. This is why the system must record the date of every purchase order and the date it was actually received. That record is what gives you each supplier's real lead time instead of the one quoted in their offer.

The third number: reorder point and safety stock

The reorder point is the balance at which a purchase order must be raised immediately. In plain words it is: what you consume during the lead time, plus a safety stock that protects you from surprises.

The first part is straightforward. If you consume ten units a day and the real lead time is fourteen days, you need one hundred and forty units to cover the waiting period. But that figure assumes everything goes to plan, and that assumption does not hold. Safety stock is added to cover two possibilities: consumption running above average, or the supplier arriving later than promised.

Safety stock is not one fixed figure applied across every item. It grows as consumption becomes more variable, as lead time becomes less predictable, and as the consequence of running out becomes more severe. A steadily consumed item from a reliable supplier needs a small buffer; an imported item with an erratic lead time whose absence stops a line deserves a much larger one.

A worked example using illustrative figures only: an item consumed at ten units a day, a real lead time of fourteen days, and a record showing delays can reach four extra days. Lead time cover is one hundred and forty units, safety stock for the possible delay is forty units, so the reorder point is one hundred and eighty units. What that means in practice: the moment the balance touches one hundred and eighty, the purchase order is raised regardless of anyone's opinion. These numbers are an illustration, not a recommendation. Every item is calculated from your own data.

Not every item deserves the same attention

A store holding a thousand items cannot be managed by reviewing each one with equal care. Ranking by importance solves this: sort your items by annual consumption value and you will find the familiar pattern that holds in most warehouses, where a small share of items accounts for the bulk of consumption value.

• The top group: a small number of high-value items. Review them regularly, count them more frequently, and calculate their reorder point carefully.
• The middle group: moderate-value items. Less frequent review and simpler rules.
• The bottom group: a large number of low-value items. Do not spend management time on them, and hold them with a comfortable buffer, because the cost of carrying them is lower than the cost of thinking about them.

Then add a second dimension on top of value: criticality. A cheap item whose absence stops the line is not a bottom-group item whatever its price, and it must be governed by top-group rules.

The other side: dead stock

Every conversation about avoiding stockouts must be matched by one about dead stock, or you end up with a full store where nothing runs out and nothing moves either. Dead stock is an item that has not moved for a long time relative to its nature, and it is not merely occupied space. It is trapped capital plus the risk of obsolescence or expiry.

The first step is a report showing the last movement date and current balance value for every item, and that is a report you cannot produce with confidence before the movement discipline of the earlier parts is in place. After that you act in order: review why the item stalled, since it may belong to a product that was discontinued; look for an alternative use inside the plant; negotiate a return or exchange with the supplier where the relationship allows it; consider selling at a discount; and finally take a documented decision to write off or dispose. What matters most is that a decision is taken, because dead stock does not resolve itself with time. It gets worse.

Then close the loop so the problem does not repeat: every item that lands on the dead stock list should have its reorder point and order quantity reviewed, because its presence there is usually the result of a wrong ordering rule rather than bad luck.

The Origami view

In the industry and warehouses sector we repeatedly see companies buy a system that has fields for reorder point and safety stock, then leave those fields empty or fill them with estimates entered once at go-live and never revisited. The result is a system producing alerts nobody trusts, so the decision falls back to the same instinct that preceded the system.

The sequence we recommend: do not calculate these numbers before you have three to six months of accurately recorded movement; start with the top group and the critical items only, not the whole store; and make reviewing the numbers a recurring task with an owner and a date rather than an event that happens once during implementation. A rule that works is a rule that gets reviewed. Anything else is just a number in a field.

In the next part

We move from quantity to identity. How do you know which supplier a material came from, which customer a product went to, and what lot number connects the two? Part five covers batch traceability and what makes a market recall possible in hours instead of days.

Sources

• Ministry of Industry and Mineral Resources — national programmes and directions on raising the efficiency of industrial facilities and supply chains.
• Saudi Authority for Industrial Cities and Technology Zones (MODON) — the industrial city environment and the logistics services tied to warehousing.
• Saudi Vision 2030 — the national direction on operational efficiency and localising supply chains.