Computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem

Jin Young Choi, Seoung Bum Kim

Research output: Contribution to journalArticle

Abstract

This paper presents a computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem by reducing the number of feature functions. The method is based on a statistical assessment of the significance of the various feature functions. This assessment can be made by combining the weighted principal components with a thresholding algorithm. The efficacy of the new feature functions selected is tested by numerical experiments. The results indicate that the feature selection method presented here can extract a small number of significant features with the potential capability of providing a compact representation of the target value function in a neuro-dynamic programming framework. Moreover, the linear parametric architecture considered holds considerable promise as a way to provide effective and computationally efficient approximations for an optimal scheduling policy that consistently outperforms the heuristics typically employed.

Original languageEnglish
Pages (from-to)2353-2362
Number of pages10
JournalInternational Journal of Production Research
Volume50
Issue number8
DOIs
Publication statusPublished - 2012 Apr 15

Fingerprint

Dynamic programming
Scheduling
Feature extraction
Approximation
Experiments

Keywords

  • capacitated re-entrant line
  • data mining
  • feature selection
  • neuro-dynamic programming
  • principal component analysis
  • scheduling

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Management Science and Operations Research
  • Strategy and Management

Cite this

@article{17888495a84e4c1fbfdca87de356febb,
title = "Computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem",
abstract = "This paper presents a computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem by reducing the number of feature functions. The method is based on a statistical assessment of the significance of the various feature functions. This assessment can be made by combining the weighted principal components with a thresholding algorithm. The efficacy of the new feature functions selected is tested by numerical experiments. The results indicate that the feature selection method presented here can extract a small number of significant features with the potential capability of providing a compact representation of the target value function in a neuro-dynamic programming framework. Moreover, the linear parametric architecture considered holds considerable promise as a way to provide effective and computationally efficient approximations for an optimal scheduling policy that consistently outperforms the heuristics typically employed.",
keywords = "capacitated re-entrant line, data mining, feature selection, neuro-dynamic programming, principal component analysis, scheduling",
author = "Choi, {Jin Young} and Kim, {Seoung Bum}",
year = "2012",
month = "4",
day = "15",
doi = "10.1080/00207543.2011.578596",
language = "English",
volume = "50",
pages = "2353--2362",
journal = "International Journal of Production Research",
issn = "0020-7543",
publisher = "Taylor and Francis Ltd.",
number = "8",

}

TY - JOUR

T1 - Computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem

AU - Choi, Jin Young

AU - Kim, Seoung Bum

PY - 2012/4/15

Y1 - 2012/4/15

N2 - This paper presents a computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem by reducing the number of feature functions. The method is based on a statistical assessment of the significance of the various feature functions. This assessment can be made by combining the weighted principal components with a thresholding algorithm. The efficacy of the new feature functions selected is tested by numerical experiments. The results indicate that the feature selection method presented here can extract a small number of significant features with the potential capability of providing a compact representation of the target value function in a neuro-dynamic programming framework. Moreover, the linear parametric architecture considered holds considerable promise as a way to provide effective and computationally efficient approximations for an optimal scheduling policy that consistently outperforms the heuristics typically employed.

AB - This paper presents a computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem by reducing the number of feature functions. The method is based on a statistical assessment of the significance of the various feature functions. This assessment can be made by combining the weighted principal components with a thresholding algorithm. The efficacy of the new feature functions selected is tested by numerical experiments. The results indicate that the feature selection method presented here can extract a small number of significant features with the potential capability of providing a compact representation of the target value function in a neuro-dynamic programming framework. Moreover, the linear parametric architecture considered holds considerable promise as a way to provide effective and computationally efficient approximations for an optimal scheduling policy that consistently outperforms the heuristics typically employed.

KW - capacitated re-entrant line

KW - data mining

KW - feature selection

KW - neuro-dynamic programming

KW - principal component analysis

KW - scheduling

UR - http://www.scopus.com/inward/record.url?scp=84861381695&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861381695&partnerID=8YFLogxK

U2 - 10.1080/00207543.2011.578596

DO - 10.1080/00207543.2011.578596

M3 - Article

VL - 50

SP - 2353

EP - 2362

JO - International Journal of Production Research

JF - International Journal of Production Research

SN - 0020-7543

IS - 8

ER -