Wilson formula history: who actually invented it
The formula is called Wilson EOQ, but the derivation predates Wilson by 21 years. Ford Whitman Harris published Q* = sqrt(2DS/H) in Factory magazine in 1913[1]. R.H. Wilson’s 1934 Harvard Business Review article popularised the formula in consulting practice, leading to the attribution as "Wilson EOQ"[2]. Erlenkotter (1990) documented the priority claim in Operations Research[3].
The Harris 1913 paper
Harris was a Westinghouse engineer who published "How Many Parts to Make at Once" in Factory, the Magazine of Management, in February 1913. The two-page article derives Q* via marginal-cost balancing without explicit calculus (Harris is satisfied with the algebraic identity Q*H/2 = DS/Q*). The formula is essentially what we use today, in identical notation.
The Wilson 1934 article
Wilson was a consultant who ran a stock-control consulting practice and published "A Scientific Routine for Stock Control" in the Harvard Business Review in 1934. The article presents the formula and a procedural framework for applying it. Wilson’s contribution is dissemination, not derivation. Because operations-research literature in the 1950s and 1960s cited Wilson’s practical framing, the formula became known as "Wilson EOQ"[2].
Erlenkotter’s 1990 priority paper
Donald Erlenkotter’s 1990 Operations Research article tracked the citation chain backward, located the Harris paper in the Factory archive, and established the priority claim formally. The community now generally credits Harris in the literature footnote but continues to call the formula "Wilson" in textbooks[3].
Why textbooks still call it Wilson
Inertia and convenience. Two generations of operations research textbooks were written under the "Wilson" convention before Erlenkotter’s priority paper appeared. Rewriting all of them is a slow process; the formula is now informally also called "Harris-Wilson" in some traditions, and in German-language literature it remains the "Andler formula" after Kurt Andler’s 1929 parallel derivation. See the Andler page for the German parallel.