Days of Supply

403. Days of Supply

How many days the current inventory will last at the current consumption rate. The “human-readable” cousin of turnover.

DOS = \frac{Average inventory}{Average daily consumption}

Equivalently:

DOS = \frac{365}{Turnover}

403.0.1. Definition variants

Variant	Formula	When to use
Unit-based	On-hand units / daily demand units	Single SKU operational view
Value-based	Inventory value / daily COGS	Aggregate financial view
Forward-looking	Inventory / forecasted daily demand	When demand changes (seasonality, promos)

The forward-looking version is more accurate for planning; the historical version is simpler and what most reports show.

403.0.2. Why DOS rather than turnover?

Turnover is a ratio (e.g., 12). DOS is a time (e.g., 30 days). Time is more intuitive for operational decisions:

“We have 45 days of supply” → vivid, actionable.
“Turnover is 8.1” → harder to feel.

Both convey the same information; just pick the one your stakeholders prefer.

403.0.3. Targets by inventory function

Different functions of inventory imply different DOS targets:

Function	Typical DOS	Reasoning
Cycle stock	$𝑄 / (2 𝑑)$	Half the cycle length; depends on EOQ
Safety stock	$𝑧 𝜎_{𝐿} / 𝑑$	Buffer for service level
Pipeline stock	$𝐿$	Lead time itself
Anticipation	varies	Days until peak demand

Total DOS = sum across functions, comparable to total inventory ÷ daily demand.

403.0.4. Reading DOS distribution

A useful diagnostic: plot DOS by SKU (histogram).

Tail of high-DOS items (e.g., > 200 days): dead stock candidates. Cross-reference with FSN — these are usually N-class (non-moving) items.
Spike at low DOS (e.g., < 7 days): chronically understocked. Check stockout rate.
Healthy bulk: cluster around the target DOS for the policy.

The aggregate company DOS hides this distribution — it’s the average. Plot it.

403.0.5. DOS vs lead time

Compare DOS to lead time:

DOS < $𝐿$ : high stockout risk. The next order won’t arrive before current stock runs out.
DOS = $𝐿$ : no buffer. Razor-thin operations.
DOS > $𝐿$ : have at least lead-time coverage; safety stock + cycle stock above $𝐿$ .

Healthy operations target $DOS \approx 1.5 \times 𝐿$ to $2 \times 𝐿$ , with the excess coming from cycle stock and modest safety stock.

Example

Given:

Average inventory: 880 units (cycle 388 + safety 30 + pipeline 462)
Daily demand: 33 units / day
Lead time: 14 days

Step 1 — total DOS

DOS = \frac{880}{33} \approx 27 days

Equivalent to turnover of $\frac{365}{27} \approx 13.5$ .

Step 2 — DOS by component

Cycle stock: $\frac{388}{33} \approx 12$ days
Safety stock: $\frac{30}{33} \approx 1$ day
Pipeline: $\frac{462}{33} = 14$ days (= $𝐿$ )
Total: $12 + 1 + 14 = 27$ days ✓

Step 3 — sanity check

Lead time = 14 days. DOS = 27 days. Buffer above lead time: 13 days (cycle + safety).
Comfortable margin. No risk of stockout from running out before next order arrives.

Step 4 — flag analysis

Suppose another SKU shows DOS = 8 days with $𝐿 = 14$ days. DOS < L: that SKU is in trouble. The next replenishment can’t arrive before current stock runs out. Stockout incoming.

Action: increase safety stock, reduce $𝑄$ for faster reorder, or expedite the inbound shipment.

Step 5 — flag dead stock

Suppose another SKU shows DOS = 365 days. That’s a year of supply on hand. Almost certainly dead stock or vastly overstocked. Action: investigate; cross-check with FSN.

DOS is the diagnostic; FSN, ABC, and stockout rate are the follow-ups.