Skip to content

Continuous Newsvendor Rule

Proposition

For continuous demand with finite mean and CDF FF, the newsvendor expected cost G(y)G(y) of ordering yy, with underage cost bb and overage cost hh, satisfies

G′(y)=(b+h)F(y)−b.G'(y)=(b+h)F(y)-b.

If FF is continuous and strictly increasing through β=b/(b+h)\beta=b/(b+h) and that quantile lies in the allowed quantity domain, the unique optimum is

y∗=F−1(β).y^*=F^{-1}(\beta).

Interpretation

The rule says that the best order is a demand level, not the average demand: order the quantity y∗y^* that demand stays at or below with probability exactly β\beta, the critical ratio. Which level is chosen depends on the costs only through β\beta.

The derivative of the expected cost plotted against the order quantity, negative below 104 and positive above it, with the optimal order quantity 104 where it crosses zero.

For meal demand uniform on [80,120][80,120] with b=3.3b=3.3 and h=2.2h=2.2, the line G′(y)=5.5F(y)−3.3G'(y)=5.5F(y)-3.3 rises from −3.3-3.3 at y=80y=80 to 2.22.2 at y=120y=120. Where it is negative, ordering a little more lowers the expected cost; where it is positive, ordering more raises it. It crosses zero at y∗=104y^*=104, where F(104)=0.6=βF(104)=0.6=\beta.

The formula for G′(y)G'(y) compares the two risks of one extra unit. It is left over with probability F(y)F(y), at cost hh, and is sold with probability 1−F(y)1-F(y), saving bb; the net effect is hF(y)−b(1−F(y))=(b+h)F(y)−bhF(y)-b(1-F(y))=(b+h)F(y)-b. Because FF never decreases, G′G' never decreases either, so GG first falls and then rises and a zero of G′G' is a global minimum. Strict increase of FF at β\beta makes the zero unique: if FF stays equal to β\beta on a whole interval, every point of that interval is optimal, so continuity alone is not enough. The quantile must also lie in the allowed range of yy; a normal quantile can be negative, and then the best order is 00.

The discrete newsvendor rule makes the same comparison for whole units, with a difference G(y+1)−G(y)G(y+1)-G(y) instead of a derivative.


Key Results

  • For uniform demand on [a,u][a,u], the optimum is a+(u−a)βa+(u-a)\beta.
  • For the course normal demand approximation N(μ,σ2)\mathcal{N}(\mu,\sigma^2), it is max⁡{0,μ+σΦ−1(β)}\max\{0,\mu+\sigma\Phi^{-1}(\beta)\}, where Φ\Phi is the standard normal CDF.
  • A continuous demand approximation may give a noninteger answer even when actual items are indivisible.

Examples

  • Meal demand uniform on [80,120][80,120] with β=0.6\beta=0.6 gives 104 meals.
  • Normal meal demand with mean 100 and standard deviation 20 gives about 105.07.

Prerequisites

Sources

Symbols: beta, mu, phi, sigma

Last updated: 20 October 2026 03:02