Sourcing with random yields and stochastic demand: A newsvendor approach

Shitao Yang, Jian Yang, Layek Abdel-Malek

Research output: Contribution to journalArticlepeer-review

84 Scopus citations


We studied a supplier selection problem, where a buyer, while facing random demand, is to decide ordering quantities from a set of suppliers with different yields and prices. We provided the mathematical formulation for the buyer's profit maximization problem and proposed a solution method based on a combination of the active set method and the Newton search procedure. Our computational study shows that the proposed method can solve the problem efficiently, and is able to generate interesting and insightful results that lead us to various managerial implications. Scope and purpose: In today's globally competitive environment, decision makers in supply chains face numerous challenges particularly regarding the selection of suppliers or outsourcing partners. To assist in this endeavor, we examined a double-layered supply chain where a buyer facing the end users has the option of selecting among a cohort of suppliers. The available suppliers may have different yield rates and unit costs. The buyer has to decide, given the stochastic nature of the problem's governing parameters, whether or not to order from each supplier, and if so how much. We developed a 'newsvendor-style' model for the problem, and proposed a solution algorithm for it. Numerical studies were performed to provide some insights for supplier selection and order quantity decisions.

Original languageEnglish (US)
Pages (from-to)3682-3690
Number of pages9
JournalComputers and Operations Research
Issue number12
StatePublished - Dec 2007

All Science Journal Classification (ASJC) codes

  • General Computer Science
  • Modeling and Simulation
  • Management Science and Operations Research


  • Newsvendor
  • Non-linear programming
  • Supplier selection
  • Supply chain


Dive into the research topics of 'Sourcing with random yields and stochastic demand: A newsvendor approach'. Together they form a unique fingerprint.

Cite this