Technical note-Multivariate partial-expectation results for exact solutions of two-stage problems

Saurabh Bansal, James S. Dyer

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We consider two-stage sequential decision-making problems where in Stage 1 an initial decision is made under a multivariate uncertainty, and in Stage 2 the uncertainty is resolved, a further decision is made based on the uncertainty realization, and the payo is observed. We focus on problems where the payo is a linear function of the multivariate uncertainty realization. Such problems can be written as single-stage nonlinear optimization problems composed of partial polyhedral expectations of the multivariate uncertainty. We identify the structural characteristics of multivariate probability density functions under which the integral expressions for the partial expectations can be directly evaluated for an exact value. We then focus on elliptical distributions, which are frequently used in operations management and do have these characteristics. We develop a sequence of three results to determine partial polyhedral expectations of elliptical probability distributions, with a special emphasis on the normal distribution. These results are useful for solving several commonly faced two-stage problems in operations management for an exact solution, performing a comparative static analysis, and rank ordering the alternatives.

Original languageEnglish (US)
Pages (from-to)1526-1534
Number of pages9
JournalOperations Research
Volume65
Issue number6
DOIs
StatePublished - Nov 1 2017

Fingerprint

Static analysis
Normal distribution
Probability distributions
Probability density function
Decision making
Uncertainty
Exact solution
Operations management
Integral
Comparative static analysis
Nonlinear optimization
Elliptical distribution
Sequential decision making
Optimization problem
Probability distribution

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Management Science and Operations Research

Cite this

@article{139519703b4344aa8b028a863ddb726c,
title = "Technical note-Multivariate partial-expectation results for exact solutions of two-stage problems",
abstract = "We consider two-stage sequential decision-making problems where in Stage 1 an initial decision is made under a multivariate uncertainty, and in Stage 2 the uncertainty is resolved, a further decision is made based on the uncertainty realization, and the payo is observed. We focus on problems where the payo is a linear function of the multivariate uncertainty realization. Such problems can be written as single-stage nonlinear optimization problems composed of partial polyhedral expectations of the multivariate uncertainty. We identify the structural characteristics of multivariate probability density functions under which the integral expressions for the partial expectations can be directly evaluated for an exact value. We then focus on elliptical distributions, which are frequently used in operations management and do have these characteristics. We develop a sequence of three results to determine partial polyhedral expectations of elliptical probability distributions, with a special emphasis on the normal distribution. These results are useful for solving several commonly faced two-stage problems in operations management for an exact solution, performing a comparative static analysis, and rank ordering the alternatives.",
author = "Saurabh Bansal and Dyer, {James S.}",
year = "2017",
month = "11",
day = "1",
doi = "10.1287/opre.2017.1604",
language = "English (US)",
volume = "65",
pages = "1526--1534",
journal = "Operations Research",
issn = "0030-364X",
publisher = "INFORMS Inst.for Operations Res.and the Management Sciences",
number = "6",

}

Technical note-Multivariate partial-expectation results for exact solutions of two-stage problems. / Bansal, Saurabh; Dyer, James S.

In: Operations Research, Vol. 65, No. 6, 01.11.2017, p. 1526-1534.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Technical note-Multivariate partial-expectation results for exact solutions of two-stage problems

AU - Bansal, Saurabh

AU - Dyer, James S.

PY - 2017/11/1

Y1 - 2017/11/1

N2 - We consider two-stage sequential decision-making problems where in Stage 1 an initial decision is made under a multivariate uncertainty, and in Stage 2 the uncertainty is resolved, a further decision is made based on the uncertainty realization, and the payo is observed. We focus on problems where the payo is a linear function of the multivariate uncertainty realization. Such problems can be written as single-stage nonlinear optimization problems composed of partial polyhedral expectations of the multivariate uncertainty. We identify the structural characteristics of multivariate probability density functions under which the integral expressions for the partial expectations can be directly evaluated for an exact value. We then focus on elliptical distributions, which are frequently used in operations management and do have these characteristics. We develop a sequence of three results to determine partial polyhedral expectations of elliptical probability distributions, with a special emphasis on the normal distribution. These results are useful for solving several commonly faced two-stage problems in operations management for an exact solution, performing a comparative static analysis, and rank ordering the alternatives.

AB - We consider two-stage sequential decision-making problems where in Stage 1 an initial decision is made under a multivariate uncertainty, and in Stage 2 the uncertainty is resolved, a further decision is made based on the uncertainty realization, and the payo is observed. We focus on problems where the payo is a linear function of the multivariate uncertainty realization. Such problems can be written as single-stage nonlinear optimization problems composed of partial polyhedral expectations of the multivariate uncertainty. We identify the structural characteristics of multivariate probability density functions under which the integral expressions for the partial expectations can be directly evaluated for an exact value. We then focus on elliptical distributions, which are frequently used in operations management and do have these characteristics. We develop a sequence of three results to determine partial polyhedral expectations of elliptical probability distributions, with a special emphasis on the normal distribution. These results are useful for solving several commonly faced two-stage problems in operations management for an exact solution, performing a comparative static analysis, and rank ordering the alternatives.

UR - http://www.scopus.com/inward/record.url?scp=85026305765&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85026305765&partnerID=8YFLogxK

U2 - 10.1287/opre.2017.1604

DO - 10.1287/opre.2017.1604

M3 - Article

VL - 65

SP - 1526

EP - 1534

JO - Operations Research

JF - Operations Research

SN - 0030-364X

IS - 6

ER -