Skip to content

Economic Order Quantity

Proposition

Under the classical EOQ assumptions with D,K,h>0D,K,h>0, the unique order quantity minimizing the total cost per unit time (TCU) TCU(y)=KD/y+hy/2TCU(y)=KD/y+hy/2 over y>0y>0 is

y∗=2KDh,t0∗=y∗D=2KDh.y^*=\sqrt{\frac{2KD}{h}},\qquad t_0^*=\frac{y^*}{D}=\sqrt{\frac{2K}{Dh}}.

The optimal policy orders y∗y^* units when stock reaches zero.


Interpretation

The economic order quantity is the batch size at which the setup cost rate KD/yKD/y and the holding cost rate hy/2hy/2 of the classical EOQ model are in balance. Larger batches cut the number of setups; smaller batches cut the average stock.

The setup cost per unit time, falling as the order size grows, the holding cost, rising linearly, and their sum, a U-shaped total cost with its minimum at the order size 22.

The setup curve 1210/y1210/y falls and the holding line 2.5y2.5y rises; they cross at y=22y=22, where each equals 5555. The total cost is lowest exactly there, at 110110 per day.

Change DD, KK and hh: the minimum of TCU(y)TCU(y) sits where the two cost rates are equal, at y∗=2KD/hy^*=\sqrt{2KD/h}.

That the minimum lies at the crossing is no coincidence. The product of the two terms, KDy⋅hy2=KDh2\frac{KD}{y}\cdot\frac{hy}{2}=\frac{KDh}{2}, does not depend on yy, and for two positive numbers a+b≥2aba+b\ge2\sqrt{ab}, since (a−b)2≥0(\sqrt{a}-\sqrt{b})^2\ge0. Hence

TCU(y)≥2KDh2=2KDh,TCU(y)\ge2\sqrt{\frac{KDh}{2}}=\sqrt{2KDh},

with equality if and only if KD/y=hy/2KD/y=hy/2, that is y2=2KD/hy^2=2KD/h. This inequality gives the same answer as setting the derivative to zero.

The square root makes the answer move slowly with the data: doubling demand raises y∗y^* only by a factor 2≈1.41\sqrt{2}\approx1.41. The cost curve is also flat near its minimum: batches of 1111 or 4444 phones instead of 2222 cost 137.5137.5 per day instead of 110110, so an order size somewhat away from y∗y^* costs little extra. The same formula reappears with lead times, is corrected by a factor with planned shortages, and serves as a candidate with price breaks.


Key Results

  • At the optimum, KD/y∗=hy∗/2KD/y^*=hy^*/2.
  • The minimum relevant cost rate is 2KDh\sqrt{2KDh}.
  • Increasing demand raises the batch size but shortens the ordering cycle.
  • The formula follows from −KD/y2+h/2=0-KD/y^2+h/2=0 and strict convexity.

Examples

  • For the illustrated phone shop, K=605K=605, D=2D=2 per day and h=5h=5 per unit per day give y∗=2⋅605⋅2/5=22y^*=\sqrt{2\cdot605\cdot2/5}=22. It orders every 22/2=1122/2=11 days; setup and holding each cost 5555 per day, giving relevant daily cost 110110.
  • The snack bar obtains y∗=2(50)(10)/(3/7)≈48.30y^*=\sqrt{2(50)(10)/(3/7)}\approx48.30 kilograms.

Prerequisites

Sources

Aliases: EOQ, EOQ formula, Wilson formula, Harris-Wilson formula, economic order quantity formula Symbols: square root

Last updated: 20 October 2026 03:02