Reorder point and safety stock
Calculates the stock level at which to reorder, how much safety stock to hold and how much to order now for an item or group.
At a glance
- What question does it answer?
- At what level to reorder, how much to hold as safety and how much to order now.
- What data is needed?
- Average weekly demand and its variability (standard deviation) from history.
- How to read the result?
- If stock plus on-order is below the reorder point the model shows how much to order up to the target level, rounded to the multiple. Compare “stock lasts” with the lead time. Example on the default values: demand is 120 units a week, spread 30, lead time 4 weeks, service level 95%: safety stock 110 units, target level 710 units. With 300 on hand and 200 on order, order 220 units (a multiple of the lot of 10); stock covers 4.2 weeks.
Why and how to reason
The reorder point solves a problem: while the delivery is on its way the item must not run out. While the order travels you sell at average demand, which is covered by “demand x lead time”. But demand fluctuates, and it is exactly the fluctuation that safety stock must cover.
Safety stock = z x variability x sqrt(lead time + period). The root, because swings of different weeks partly cancel: over 5 weeks the spread grows not 5 times but sqrt(5), about 2.24. The coefficient z sets the service level: for 95% it is 1.645, meaning you accept running short in 5 cases out of 100.
Example. Demand 120 a week, variability 30, lead time 4 weeks, review period 1. Horizon is 5 weeks: safety stock 1.645 x 30 x 2.236, about 110. Target level 120 x 5 + 110 = 710. On hand 300, on order 200: order 210, which rounded to the lot of 10 is 220.
When to use it
It is meant for items with stable demand: basics and consumables that must be in stock all year. It is unsuitable for new items without history and season hits: their demand shifts, so use the nearest weeks. Assumptions: demand is normal and independent week to week, and the lead time is stable. If the supplier misses dates, increase lead-time variability, not only demand variability.
How to read the result
Safety stock is the price of reliability: raising service from 95 to 99% will increase it noticeably, and cash frozen in stock grows. Compare “stock lasts for” with the lead time: 4.2 weeks of cover against a 4-week delivery means you are on the edge. The “order now” amount accounts for what is already in transit, so do not double the order. Decision: order immediately if the recommendation is above zero; if zero, check the dates. Then judge whether safety stock is too large on expensive items. If 95% service gives a safety stock that seems excessive, check demand variability: it is often inflated by one-off promotions in the history.
Common mistakes
The usual mistakes: entering weekly demand with the lead time in days; not including what is already ordered; using mean deviation where the standard deviation is needed; setting a lot size larger than reasonable. Also: applying the formula to a seasonal item in the middle of its season. Also: taking variability from a short series of three weeks when at least ten are needed.
What data is needed?
- Average weekly demand and its variability (standard deviation) from history.
- Lead time, review period, service level, stock and on-order quantity.
Inputs
| Field | Unit | Value in the example | Meaning |
|---|---|---|---|
| How you reorder | On a schedule: I check stock every period and order up to a target level | The choice changes the safety stock: with scheduled orders it covers the lead time plus the review period, with stock-level orders only the lead time. Only the metrics of the chosen scheme are shown. | |
| Average weekly demand | units/week | 120 | Average units sold per week; take the average of the last 8–12 weeks excluding stock-out periods. |
| Demand variability (std. deviation) | units/week | 30 | How much weekly demand deviates from the average. Take it from sales history; without history, 20–40% of average demand is only a rough guess, not a standard. |
| Lead time | weeks | 4 | Weeks from placing the order to the goods arriving in the warehouse, including production and delivery. |
| Review period | weeks | 1 | How often you place orders. The less often, the more stock you need. |
| Service level | 95% | The probability that the item is in stock. The higher it is, the more stock and cost. | |
| Stock on hand | units | 300 | Free stock in the warehouse and stores now, excluding defects and reservations. |
| On order | units | 200 | Already ordered and not yet received: in transit, in production. |
| Order multiple | units | 10 | Minimum lot or pack: the order is rounded up to a multiple. |
How it is calculated
| Step | Formula |
|---|---|
| Safety stock | z × variability × √(lead time + review) |
| Order-up-to level | Demand × (lead + review) + safety stock |
| Order | Target − on hand − on order, rounded up to the multiple |
Worked numbers
Default values from the app; in the interactive version you enter your own.
- Safety stock
- 110 unitsStock for demand spikes or delivery delays.
- Order-up-to level
- 710 unitsDemand over lead time and review period plus safety stock: you top stock up to it at every check.
- Order now
- 220 unitsUp to the target level, rounded to the multiple.
- Stock lasts
- 4.2 weeks(On hand + on order) ÷ weekly demand.
On hand plus on order (500 units) is below the target 710 units: order 220 units.
Nuances and limits
- Demand is assumed normal and independent by week; for seasonal bestsellers use the nearest weeks.
- Raising the service level from 95 to 99% noticeably increases stock.