Reorder Point Formula: Safety Stock for Real Lead Times

The reorder point formula is reorder point = (average daily demand × lead time) + safety stock. Place a purchase order (PO) when your inventory position, meaning stock on hand plus stock already ordered but not yet received, falls to that number. Lead time is the days from placing the PO to receiving the goods.

Take an item that sells 50 units a day with a 30-day lead time. Lead-time demand, the stock you expect to sell while waiting, is 50 × 30 = 1,500 units. Safety stock is the extra you hold for days when demand or delivery runs worse than average. For this item the next section calculates 842 units, so the reorder point is 1,500 + 842 = 2,342 units.

Three mistakes cause most bad reorder points:

• Mixed units. Daily demand needs lead time in days. Convert weeks before you multiply.
• Checking on-hand stock only. An order already on the water counts. Otherwise you reorder twice for the same gap.
• Confusing when with how much. The reorder point says when to order. The reorder quantity says how much.

Hold enough safety stock to cover swings in both daily demand and lead time at your chosen service level: safety stock = Z × √(L × σd² + d² × σL²). For a 95% service level, Z is 1.645.

The terms:

• d is average daily demand. L is average lead time in days.
• σd and σL (sigma) are the standard deviations of daily demand and of lead time. Each is the typical size of the swing around the average, and =STDEV.S() calculates it in a spreadsheet.
• Service level is the share of orders that arrive before you run out. Z is the factor that turns a service level into a number of standard deviations. Safety stock is conventionally sized with Z = 1.645, often rounded to 1.65, for 95%. In a sheet, =NORM.S.INV(0.95) returns it, and =NORM.S.INV(0.98) returns about 2.05 for 98%. The combined formula for demand and lead-time variability is the standard result in Silver, Pyke and Thomas, Inventory and Production Management in Supply Chains.

Worked example: d = 50, σd = 20, L = 30, σL = 10.

• L × σd² = 30 × 400 = 12,000
• d² × σL² = 2,500 × 100 = 250,000
• The square root of 262,000 is about 512. Multiply by 1.645 to get 842 units.

Our recommendation is 95% for ordinary lines. Use 98% only where a stockout stops a customer order or a production line, because every step up in service level adds stock. The formula assumes demand and lead time vary independently around steady averages. For seasonal items, calculate d and σd from the same season, not the last 90 days.

The shortcut safety stock = Z × σd × √L counts demand swings only. When delivery times also swing, it leaves orders short. In our model, with a 10-day standard deviation in lead time, only about 48% of orders arrived without a stockout, against about 92% for the full formula.

Our model. We simulated one item: demand averaging 50 units a day (standard deviation 20, uneven day to day, never below zero), lead time averaging 30 days drawn from a normal curve and rounded to whole days, and orders of 1,500 units placed when inventory position reached the reorder point. Z = 1.645. Unmet demand is treated as lost sales, not backordered. We ran 60 simulated runs of 1,500 days per setting, in two separate random seeds, roughly 3,000 orders per setting. A stockout was any unmet demand while an order was open. Each cell shows the range across the two seeds. This is our model, not field data.

With no safety stock, the reorder point is only the average, and roughly half of orders fall short even when deliveries are perfectly regular. The demand-only shortcut works until lead time varies, then it collapses. The full formula lands near 93% rather than 95% because the order triggers after the day's demand has already taken the inventory position below the reorder point, so the buffer is slightly smaller than the formula assumes.

The shortfalls were small. At the 10-day setting, the demand-only version still filled about 94% of units demanded, but a stockout touched about 52% of orders. The full formula filled about 99.5% of units.

The cost is stock. At a 10-day standard deviation, safety stock rises from 180 to 842 units, nearly five times as much. For illustration, at a landed cost of ₹400 a unit, the extra 662 units tie up ₹2,64,800. Real shipment delays tend to run late rather than evenly around the average, so for imported lines we would treat the full formula as the minimum.

Use the lead time you measure yourself: days from placing the PO to receiving the goods, for your last 5 to 10 orders. Take the average as L and the standard deviation as σL. A supplier's quoted lead time leaves out customs, haulage and delays.

The port you use changes the wait. The National Time Release Study, run by the Central Board of Indirect Taxes and Customs (CBIC), measures customs release time from a sample of Bills of Entry, the import declarations filed with customs, taken in the first week of January each year. The 2025 edition, hosted by the Press Information Bureau (PIB), sets out that method on page 26. The 2026 port figures, as reported by India Shipping News, are below, with 2025 from the 2025 report (pages 30 and 31). Figures as of October 2026:

At Mundra, 78.39% of Bills of Entry met the 48-hour target in 2026. Kolkata's 2026 average is 122 hours 42 minutes, about 123 hours or 5 days, longer than Mundra's, from customs release alone. Sailing, port handling, trucking and supplier production time come on top, which is why measuring from your own POs is more reliable than adding up stages. With fewer than five deliveries on record, use your slowest delivery so far as L and set σL to 0, because the slowest delivery already builds in the delay. Update both after each new delivery.

Reorder level and reorder point are the same thing: the stock level that triggers an order. The minimum level is the buffer you should not dip below, which is safety stock. Reorder quantity is how much to order. The syllabus formulas below differ from the practical one.

The reorder level, maximum level and danger level formulas appear in exam notes hosted by the Institute of Chartered Accountants of India (ICAI). The same notes work a practice question: a maximum level of 8,000 kg, a reorder quantity of 5,000 kg and minimum usage of 50 kg an hour over 8 hours a day for 4 days gives a reorder level of 8,000 − (5,000 − 50 × 8 × 4) = 4,600 kg.

In our illustration (invented figures, not from the ICAI notes), normal usage is 300 units a week, maximum 450, minimum 150, and the re-order period is 4 to 6 weeks. The reorder level is 450 × 6 = 2,700 units. Lead-time demand at the normal 5-week period is 300 × 5 = 1,500 units, so the minimum level is 2,700 − 1,500 = 1,200 units, which is 80% of lead-time demand held as buffer. The textbook method assumes the busiest week meets the slowest delivery every time. Use it in an exam. For stock, we recommend the reorder point formula. For items where a stockout is unaffordable, raise the service level to 98% rather than switching to the worst-case method.

Build one row per item with ten columns: demand, demand swing, lead time, lead-time swing, Z, lead-time demand, safety stock, reorder point, stock position and an order flag. Every column is a formula or a number you measure.

Run these checks before you trust the sheet:

• Self-test. Enter B = 50, C = 20, D = 30, E = 10, F = 1.645. Pass: G = 1,500, H shows 842 and I = 2,342. Fail: any other value, so recheck the brackets in H.
• Units. Pass: demand per day and lead time in days in every row. Fail: any lead time in weeks.
• Lead-time sample. Pass: at least 5 POs behind D and E. Fail: fewer, so use your slowest delivery as D and enter 0 in E, because the slowest delivery already builds in the delay.
• Trigger. Pick an item flagged "Order". Pass: it has no open PO, or you raise one today. Fail: you find an open PO that the sheet ignored, so add it to column J.
• Review. Check the sheet's last-updated date. Pass: updated within the last 30 days and after any delivery later than L + σL. Fail: older, so recalculate columns B to I, because demand and supplier performance both drift. Managing warehouse inventory without stock surprises covers the weekly routine that keeps the inputs honest.

A sheet works well for a few dozen items. Once you hold hundreds of items across several locations, see inventory software for small businesses.

Enter each item's reorder point as its minimum stock level in Arka Inventory. Our Inventory Alerts then flag items approaching or falling below minimum stock levels, and items under back order, so your team sees what needs attention without checking a spreadsheet row by row.

An alert is a prompt to check what is already on order before you raise another PO, which is the inventory position test from the sheet. When a delivery runs late, recalculate the reorder point in your sheet and enter the new figure as the item's minimum stock level in Arka Inventory.

To plan purchases from real orders, forecasts and stock, use our Forecasting and Purchase Automation feature. Arka Inventory is built natively on Salesforce, so purchasing sits alongside your sales and service data. To tie safety stock to cash, measure how fast stock turns with inventory turnover days. More safety stock lengthens the days.

Arka Inventory's Basic plan costs $199 a month, paid yearly, as of October 2026. See the plans or book a demo with our sales team.