eng Science and Education Publishing American Journal of Applied Mathematics and Statistics 2328-7292 2024-04-21 12 2 24 27 10.12691/ajams-12-2-1 AJAMS20241221 article Quan Yuan qyuan@bsu.edu 1 Zhixin Yang 1 Yayuan Xiao 1 Department of Mathematical Sciences, Ball State University, Muncie IN, USA This paper investigates the application of the Kiefer-Wolfowitz (KW) algorithm, a stochastic approximation technique, to solve the newsvendor problem under uncertain demand. The proposed approach enables the newsvendor to learn from observed profits and converge to the optimal order quantity, even when the demand distribution is unknown. Numerical experiments demonstrate the algorithm's effectiveness in handling stochastic demand and provide insights into its convergence properties. The paper highlights the potential of stochastic approximation methods in tackling inventory management challenges and discusses future research directions. https://pubs.sciepub.com/ajams/12/2/1/ajams-12-2-1.pdf Newsvendor problem stochastic approximation Kiefer-Wolfowitz algorithm