Reorder point and safety stock calculator
Your data
The two variation fields are what separate this from a rule of thumb. If you leave both at zero the answer is simply demand times lead time, and a stockout every second cycle is the price of that simplicity.
Service level here means the share of cycles that end without running out, not the share of orders you fill. They are different numbers and the second one is always the kinder of the two.
Results
Order again when stock falls to this many units
—
| What you expect to use while you wait | — |
| Safety stock on top of that | — |
| Standard deviations that service level asks for | — |
| If the lead time were halved, the safety stock would fall by this many in a hundred | — |
| Share of the uncertainty that comes from the lead time, in percent | — |
What each extra bit of service level costs
| Service level | Deviations from the mean | Safety stock | Added since the line above |
|---|
Read the last column downwards and the shape of the problem appears. Going from ninety-five to ninety-nine costs less than the four-tenths of a point that follows it. The last stretch is the expensive one, which is why aiming at never running out is a decision about money rather than about service.
The other thing worth carrying away is that safety stock follows the square root of the lead time, not the lead time itself. When the lead time is steady, halving the wait does not halve the stock you have to hold: it takes off about twenty-nine in a hundred. When the lead time also varies, shortening it helps even less than that, and the row above works out the figure for your own numbers rather than quoting the twenty-nine. That is the same square root telling you that a supplier who is merely slow is easier to live with than one who is unpredictable.
What is the difference between the reorder point and the economic order quantity?
They answer different halves of the same policy. The economic order quantity says how much to buy each time; the reorder point says when to buy it. You need both, and neither one replaces the other.
In practice the order quantity is a cost question, balancing what it costs to place an order against what it costs to hold stock. The reorder point is a risk question: how often you are willing to run out.
If my supplier halves the lead time, does my safety stock halve too?
No, and this is the part almost everyone gets wrong. Safety stock follows the square root of the lead time, so halving the wait removes about 29% of it, not 50%.
The same square root explains something else worth knowing: a supplier who is slow but predictable is much easier to live with than one whose lead time swings. Steadiness is worth more than speed here.
Why does the last bit of service level cost so much?
Because the normal curve thins out at the edges. Getting from 50% to 95% takes 1.645 standard deviations. The next four points, from 95% to 99%, cost another 0.681. The nine-tenths of a point after that, from 99% to 99.9%, costs 0.764 more, which is more than the four points before it.
So aiming at never running out is a decision about money, not about service. The table on the page prints what each step adds in units of your own stock.
| Service level | Deviations from the mean | Added since the line above |
|---|---|---|
| 50% | 0.000 | — |
| 90% | 1.282 | 1.282 |
| 95% | 1.645 | 0.363 |
| 99% | 2.326 | 0.681 |
| 99.9% | 3.090 | 0.764 |
Which matters more, uncertain demand or an uncertain lead time?
Usually the lead time, and by more than people expect, because the average demand enters that term squared. On a fast-moving item, half a day of uncertainty in delivery can outweigh all the day-to-day swing in sales.
The page prints which of the two is the larger share of your uncertainty. When it says the lead time, arguing with the supplier will do more for your stock than any amount of better forecasting.
Reactions
0
0 Comments
Be the first to comment