Testing Solution Quality in Stochastic Programming: a Single Replication Procedure

Abstract

We develop a procedure for testing the quality of a candidate solution for a class of stochastic programs. Quality is defined via the optimality gap and the procedure’s output is a confidence interval on this gap. We review a multiple-replications procedure and then, we present a result that justifies a computationally simplified single-replication… (More)

Topics

Cite this paper

@inproceedings{Bayraksan2004TestingSQ, title={Testing Solution Quality in Stochastic Programming: a Single Replication Procedure}, author={G{\"{u}zin Bayraksan and David P. Morton}, year={2004} }