The sequential stochastic assignment problem with random success rates

Arash Khatibi, Golshid Baharian, Estelle R. Kone, Sheldon H. Jacobson

Research output: Contribution to journalArticlepeer-review


Given a finite number of workers with constant success rates, the Sequential Stochastic Assignment problem (SSAP) assigns the workers to sequentially arriving tasks with independent and identically distributed reward values, so as to maximize the total expected reward. This article studies the SSAP, with some (or all) workers having random success rates that are assumed to be independent but not necessarily identically distributed. Several assignment policies are proposed to address different levels of uncertainty in the success rates. Specifically, if the probability density functions of the random success rates are known, an optimal mixed policy is provided. If only the expected values of these rates are known, an optimal expectation policy is derived. © 2014

Original languageEnglish (US)
Pages (from-to)1169-1180
Number of pages12
JournalIIE Transactions (Institute of Industrial Engineers)
Issue number11
StatePublished - Nov 2 2014


  • Sequential decision-making
  • assignment policy
  • stochastic processes

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering


Dive into the research topics of 'The sequential stochastic assignment problem with random success rates'. Together they form a unique fingerprint.

Cite this