Feedback control for priority rules in re-entrant semiconductor manufacturing

Fenglan He, Hans Armbruster, Michael Herty, Ming Dong

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Typical semiconductor production is re-entrant and hence requires priority decisions when parts compete for production capacity at the same machine. A standard way to run such a factory is to start to plan and to finish according to demand. Often this results in a push policy where early production steps have priority over later production steps at the beginning of the production line and a pull policy where later steps have priority at the end of the production line. The point where the policies switch is called the push-pull-point (PPP). We develop a control scheme based on moving the PPP in a continuum model of the production flow. We show that this control scheme significantly reduces the mismatch between demand and production output. The success of the control scheme as a function of the frequency of control action is analyzed and optimal times between control actions are determined.

Original languageEnglish (US)
Pages (from-to)4655-4664
Number of pages10
JournalApplied Mathematical Modelling
Volume39
Issue number16
DOIs
StatePublished - Aug 15 2015

Fingerprint

Priority Rules
Semiconductor Manufacturing
Feedback Control
Feedback control
Semiconductor materials
Production Line
Continuum Model
Semiconductors
Switch
Industrial plants
Output
Switches
Policy

Keywords

  • Dispatch policy control
  • Feedback control
  • Partial differential equations
  • Re-entrant production

ASJC Scopus subject areas

  • Applied Mathematics
  • Modeling and Simulation

Cite this

Feedback control for priority rules in re-entrant semiconductor manufacturing. / He, Fenglan; Armbruster, Hans; Herty, Michael; Dong, Ming.

In: Applied Mathematical Modelling, Vol. 39, No. 16, 15.08.2015, p. 4655-4664.

Research output: Contribution to journalArticle

He, Fenglan ; Armbruster, Hans ; Herty, Michael ; Dong, Ming. / Feedback control for priority rules in re-entrant semiconductor manufacturing. In: Applied Mathematical Modelling. 2015 ; Vol. 39, No. 16. pp. 4655-4664.
@article{672fa930476f4422b269a348284cb8a5,
title = "Feedback control for priority rules in re-entrant semiconductor manufacturing",
abstract = "Typical semiconductor production is re-entrant and hence requires priority decisions when parts compete for production capacity at the same machine. A standard way to run such a factory is to start to plan and to finish according to demand. Often this results in a push policy where early production steps have priority over later production steps at the beginning of the production line and a pull policy where later steps have priority at the end of the production line. The point where the policies switch is called the push-pull-point (PPP). We develop a control scheme based on moving the PPP in a continuum model of the production flow. We show that this control scheme significantly reduces the mismatch between demand and production output. The success of the control scheme as a function of the frequency of control action is analyzed and optimal times between control actions are determined.",
keywords = "Dispatch policy control, Feedback control, Partial differential equations, Re-entrant production",
author = "Fenglan He and Hans Armbruster and Michael Herty and Ming Dong",
year = "2015",
month = "8",
day = "15",
doi = "10.1016/j.apm.2015.03.061",
language = "English (US)",
volume = "39",
pages = "4655--4664",
journal = "Applied Mathematical Modelling",
issn = "0307-904X",
publisher = "Elsevier Inc.",
number = "16",

}

TY - JOUR

T1 - Feedback control for priority rules in re-entrant semiconductor manufacturing

AU - He, Fenglan

AU - Armbruster, Hans

AU - Herty, Michael

AU - Dong, Ming

PY - 2015/8/15

Y1 - 2015/8/15

N2 - Typical semiconductor production is re-entrant and hence requires priority decisions when parts compete for production capacity at the same machine. A standard way to run such a factory is to start to plan and to finish according to demand. Often this results in a push policy where early production steps have priority over later production steps at the beginning of the production line and a pull policy where later steps have priority at the end of the production line. The point where the policies switch is called the push-pull-point (PPP). We develop a control scheme based on moving the PPP in a continuum model of the production flow. We show that this control scheme significantly reduces the mismatch between demand and production output. The success of the control scheme as a function of the frequency of control action is analyzed and optimal times between control actions are determined.

AB - Typical semiconductor production is re-entrant and hence requires priority decisions when parts compete for production capacity at the same machine. A standard way to run such a factory is to start to plan and to finish according to demand. Often this results in a push policy where early production steps have priority over later production steps at the beginning of the production line and a pull policy where later steps have priority at the end of the production line. The point where the policies switch is called the push-pull-point (PPP). We develop a control scheme based on moving the PPP in a continuum model of the production flow. We show that this control scheme significantly reduces the mismatch between demand and production output. The success of the control scheme as a function of the frequency of control action is analyzed and optimal times between control actions are determined.

KW - Dispatch policy control

KW - Feedback control

KW - Partial differential equations

KW - Re-entrant production

UR - http://www.scopus.com/inward/record.url?scp=84930758036&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930758036&partnerID=8YFLogxK

U2 - 10.1016/j.apm.2015.03.061

DO - 10.1016/j.apm.2015.03.061

M3 - Article

VL - 39

SP - 4655

EP - 4664

JO - Applied Mathematical Modelling

JF - Applied Mathematical Modelling

SN - 0307-904X

IS - 16

ER -