Why Inventory Management Matters
Inventory is simultaneously an asset and a liability. Too little, and you suffer stockouts, lost sales, and idle production lines. Too much, and you tie up cash, consume warehouse space, and risk obsolescence. Inventory management is the discipline of finding the right balance: how much to order and when to order it. The classic models below answer those two questions precisely.
The Holding vs. Ordering Cost Trade-off
Two opposing costs drive order-size decisions:
- Ordering cost (S): the fixed cost incurred each time you place an order — paperwork, setup, shipping, receiving. It is independent of order size, so ordering in large batches spreads it thin.
- Holding (carrying) cost (H): the annual cost of keeping one unit in stock — capital tied up, storage, insurance, spoilage. Larger orders raise average inventory and therefore holding cost.
Order in tiny batches and you pay for many orders; order in huge batches and you pay to store a mountain of inventory. The sweet spot between them is the Economic Order Quantity.
The EOQ Formula and Its Derivation
The total annual cost is the sum of ordering and holding cost:
TC(Q) = (D/Q)·S + (Q/2)·H
where D is annual demand, Q is order quantity, S is cost per order, and H is annual holding cost per unit. The first term is annual ordering cost (number of orders × cost per order); the second is annual holding cost (average inventory Q/2 × holding cost per unit). To minimize, take the derivative with respect to Q and set it to zero:
dTC/dQ = −(D·S)/Q² + H/2 = 0 → Q² = 2DS/H
EOQ = √(2DS ÷ H)
An elegant property falls out: at the EOQ, annual ordering cost exactly equals annual holding cost. The total cost curve is also fairly flat near the optimum, so being modestly off the exact EOQ has only a small cost penalty — a comforting fact when inputs are uncertain.
A Worked Example
Suppose a distributor has the following data for one product:
- Annual demand D = 12,000 units
- Ordering cost S = $50 per order
- Holding cost H = $3 per unit per year
EOQ = √(2 × 12,000 × 50 ÷ 3) = √(1,200,000 ÷ 3) = √400,000 ≈ 632 units
Check the balance at Q = 632:
- Annual ordering cost = (12,000 ÷ 632) × $50 ≈ 18.99 × $50 ≈ $949
- Annual holding cost = (632 ÷ 2) × $3 = 316 × $3 = $948
The two costs are equal (to rounding), confirming the optimum. The number of orders per year is about 19, and the total relevant annual cost is roughly $1,897. You can run your own numbers on the EOQ calculator.
Reorder Point and Lead Time
EOQ tells you how much; the reorder point tells you when. The reorder point (ROP) is the inventory level that triggers a new order, set so that stock arrives just as you run out:
ROP = average daily demand × lead time (days)
Continuing the example, if demand is steady at 12,000 ÷ 300 working days = 40 units/day and lead time is 5 days, then ROP = 40 × 5 = 200 units. When inventory drops to 200, place an order for the EOQ of 632.
Safety Stock for Uncertainty
Real demand and lead times vary, so the simple ROP risks frequent stockouts. Safety stock is a buffer against that variability:
Safety stock = Z × σLT
where Z is the service-level factor from the standard normal distribution and σLT is the standard deviation of demand over the lead time. The ROP then becomes ROP = (demand during lead time) + safety stock. Common Z values:
| Service Level | Z factor | Interpretation |
|---|---|---|
| 90% | 1.28 | Stockout in 1 of 10 cycles |
| 95% | 1.65 | Stockout in 1 of 20 cycles |
| 98% | 2.05 | Stockout in 1 of 50 cycles |
| 99% | 2.33 | Stockout in 1 of 100 cycles |
Note the steep cost: pushing service level from 95% toward 100% requires disproportionately more safety stock, which is why few items justify a near-perfect target.
ABC Analysis: Focusing Effort
Not all inventory deserves equal attention. ABC analysis applies the Pareto principle by ranking items by annual dollar usage (unit cost × annual demand):
- A items — roughly the top 20% of items by value, accounting for about 70–80% of total inventory value. Tight control: frequent review, accurate forecasts, low safety stock relative to value.
- B items — the middle tier, moderate value and moderate control.
- C items — the bottom 50%+ of items that make up only a small share of value. Manage with simple rules (large safety stock, infrequent review) because the carrying cost is trivial.
Spending the most management attention on A items, where the dollars are, yields the highest payoff per hour of effort.
Putting It Together
A complete policy combines all four ideas: classify items with ABC, set order size with EOQ, set the trigger with the reorder point, and buffer uncertainty with safety stock sized to a sensible service level. These models assume relatively stable demand; for highly variable or lumpy demand, more advanced approaches (periodic review, forecasting, or lean pull systems) may serve better. Inventory turns — annual demand ÷ average inventory — is a useful overall scorecard for whether your policies are keeping inventory lean without starving the operation.