What this tool does
A reorder point is the stock level at which placing an order beats running out, and it has two parts: the demand you expect to sell while the next shipment is still in transit, and a safety buffer to cover the days your forecast was wrong or your supplier was late. Get it wrong one way and you park cash in slow-moving stock; get it wrong the other and a product you sell every day goes out of stock. This calculator derives both figures from lead time and daily usage, and gives you the order-by date rather than a number to interpret.
How it works
Lead-time demand is the simple half: average daily demand multiplied by the supplier's lead time in days. Forty units a day on a seven-day lead time is 280 units, and that is what leaves the shelf between placing the order and the pallet arriving. Lengthen the lead time and the reorder point moves immediately, which is why a supplier's reliability shows up in your inventory number before it ever shows up in your cash.
Safety stock is the buffer, and how you set it depends on how much you are prepared to assume. Where you have enough history to measure variability, safety stock is z multiplied by the standard deviation of daily demand multiplied by the square root of lead time. The square root is the part people miss: demand uncertainty grows with the root of the window rather than in proportion to it, so a fourteen-day lead time needs about 1.41 times the buffer of a seven-day one, not twice. The z value comes from your target service level, 1.645 for 95% and 2.05 for 98%, and it is a policy choice rather than a measurement.
The economic order quantity answers a different question. EOQ is the square root of two times annual demand times ordering cost, divided by unit cost times the holding rate, and it balances the cost of raising a purchase order against the cost of carrying the stock. It is worth calculating because a reorder point tells you when to fire and EOQ tells you how much. Reordering five units at a time on 2,000 units of annual demand means raising purchase orders far more often than the economics justify, and each one has a fixed cost attached to it.
Keep the honest caveat in front of you: this is a demand model, not a forecast. It assumes the future resembles the past, and for a product with seasonality, a trend, a promotion running or a new entrant, that is the weaker of the two assumptions on which the whole calculation rests. Inventory turns, meaning cost of goods sold divided by average inventory value, is the figure that tells you whether the resulting policy is healthy, and it is the one to watch once the reorder point is live in your system.
Worked example
Average daily demand of 40 units with a standard deviation of 12 units a day, a 7-day lead time, a 95% service level (z = 1.645), 1,200 units a month, a 900 rupee purchase order cost and a 250 rupee unit cost held at 25% a year.
- Lead-time demand = 40 x 7 = 280 units
- Safety stock = 1.645 x 12 x sqrt(7) = 1.645 x 31.75 = 52.2, rounded up to 53 units
- Reorder point = 280 + 53 = 333 units
- Days of cover at that level = 333 / 40 = 8.3, so reorder 8 days from now
- Annual demand = 1,200 x 12 = 14,400 units; holding cost = 25% x 250 = 62.50 per unit per year
- EOQ = sqrt(2 x 14,400 x 900 / 62.50) = sqrt(414,720) = 644 units, about 22 orders a year
- Average inventory = 644 / 2 + 53 = 375 units, so 36,00,000 / 93,750 = 38 turns a year
Reorder at 333 units, which is 8 days of cover away, and buy 644 units at a time. At 38 turns a year the stock cycles roughly every ten days, so a supplier delay is a bigger risk to this item than a demand error.
Accuracy and limitations
- This is a demand model built on past usage, not a forecast. Seasonality, a trend, a promotion or a new competitor will all break the assumption that next month resembles last month, and the tool has no way to see any of them coming.
- Supplier reliability outweighs the arithmetic. A buffer sized for a 95% service level against demand still runs out if the lead time itself slips, so track actual lead times rather than the quoted ones and re-run this when they change.
- The holding cost is a rate rather than an invoice. If you do not know your carrying cost, treat safety stock as a policy percentage of lead-time demand instead of implying a z-score precision the input data does not support.
Frequently asked questions
- How do I calculate a reorder point?
- Add lead-time demand to safety stock. Lead-time demand is average daily demand times lead time in days, so 40 units a day on a 7-day lead time is 280 units. Add 53 units of safety stock and the reorder point is 333 units.
- What is safety stock and how much should I hold?
- It is the buffer that absorbs a demand forecast being wrong or a delivery running late. Derived from your own history it is z times the standard deviation of daily demand times the square root of lead time, where z is your chosen service level: 1.645 for 95%, 2.05 for 98%.
- What is the economic order quantity?
- EOQ is the square root of two times annual demand times order cost, divided by unit cost times the annual holding rate. It is the order size that minimises ordering and holding costs together, and it is worth using instead of raising a small purchase order every time stock dips.
- How often should I reorder stock?
- Divide annual demand by the order quantity to get the number of orders a year, then spread that across twelve months. At 14,400 units a year and a 644-unit order quantity that is about 22 orders, roughly one every sixteen days, with safety stock covering the gap between them.
- Why did I run out of stock even with safety stock?
- Almost always because the lead time moved, not because the demand model was wrong. A supplier that delivers in 10 days rather than 7 adds three days of demand on top of everything you already allowed for, and no amount of demand forecasting protects against that.
- How many days of stock should a shop keep?
- Enough to cover the lead time plus the buffer you have decided to pay for, and no more. Run the number, then sanity-check it against inventory turns, which is cost of goods sold divided by average inventory value, because a policy implying 40 turns a year is wrong for most categories.