Demand signal modelling: A short-range panel forecasting algorithm for semiconductor firm device-level demand

Russell J. Elias, Douglas Montgomery, Stuart A. Low, Murat Kulahci

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

A model-based approach to the forecasting of short-range product demand within the semiconductor industry is presented. Device-level forecast models are developed via a novel two-stage stochastic algorithm that permits leading indicators to be optimally blended with smoothed estimates of unit-level demand. Leading indicators include backlog, bookings, delinquencies, inventory positions, and distributor resales. Group level forecasts are easily obtained through upwards aggregation of the device level forecasts. The forecasting algorithm is demonstrated at two major US-based semiconductor manufacturers. The first application involves a product family consisting of 254 individual devices with a 26-month training dataset and eight-month ex situ validation dataset. A subsequent demonstration refines the approach, and is demonstrated across a panel of six high volume devices with a 29-month training dataset and a 13-month ex situ validation dataset. In both implementations, significant improvement is realised versus legacy forecasting systems.

Original languageEnglish (US)
Pages (from-to)253-278
Number of pages26
JournalEuropean Journal of Industrial Engineering
Volume2
Issue number3
DOIs
StatePublished - Mar 2008

Fingerprint

Semiconductor devices
Semiconductor materials
Demonstrations
Agglomeration
Industry

Keywords

  • Demand forecasting
  • Enterprise resource planning
  • Operations modelling
  • Semiconductor fabrication
  • Statistical forecasting
  • Supply chain management

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Demand signal modelling : A short-range panel forecasting algorithm for semiconductor firm device-level demand. / Elias, Russell J.; Montgomery, Douglas; Low, Stuart A.; Kulahci, Murat.

In: European Journal of Industrial Engineering, Vol. 2, No. 3, 03.2008, p. 253-278.

Research output: Contribution to journalArticle

@article{8e32641c0d4c42438050f720167598ee,
title = "Demand signal modelling: A short-range panel forecasting algorithm for semiconductor firm device-level demand",
abstract = "A model-based approach to the forecasting of short-range product demand within the semiconductor industry is presented. Device-level forecast models are developed via a novel two-stage stochastic algorithm that permits leading indicators to be optimally blended with smoothed estimates of unit-level demand. Leading indicators include backlog, bookings, delinquencies, inventory positions, and distributor resales. Group level forecasts are easily obtained through upwards aggregation of the device level forecasts. The forecasting algorithm is demonstrated at two major US-based semiconductor manufacturers. The first application involves a product family consisting of 254 individual devices with a 26-month training dataset and eight-month ex situ validation dataset. A subsequent demonstration refines the approach, and is demonstrated across a panel of six high volume devices with a 29-month training dataset and a 13-month ex situ validation dataset. In both implementations, significant improvement is realised versus legacy forecasting systems.",
keywords = "Demand forecasting, Enterprise resource planning, Operations modelling, Semiconductor fabrication, Statistical forecasting, Supply chain management",
author = "Elias, {Russell J.} and Douglas Montgomery and Low, {Stuart A.} and Murat Kulahci",
year = "2008",
month = "3",
doi = "10.1504/EJIE.2008.017686",
language = "English (US)",
volume = "2",
pages = "253--278",
journal = "European Journal of Industrial Engineering",
issn = "1751-5254",
publisher = "Inderscience Enterprises Ltd",
number = "3",

}

TY - JOUR

T1 - Demand signal modelling

T2 - A short-range panel forecasting algorithm for semiconductor firm device-level demand

AU - Elias, Russell J.

AU - Montgomery, Douglas

AU - Low, Stuart A.

AU - Kulahci, Murat

PY - 2008/3

Y1 - 2008/3

N2 - A model-based approach to the forecasting of short-range product demand within the semiconductor industry is presented. Device-level forecast models are developed via a novel two-stage stochastic algorithm that permits leading indicators to be optimally blended with smoothed estimates of unit-level demand. Leading indicators include backlog, bookings, delinquencies, inventory positions, and distributor resales. Group level forecasts are easily obtained through upwards aggregation of the device level forecasts. The forecasting algorithm is demonstrated at two major US-based semiconductor manufacturers. The first application involves a product family consisting of 254 individual devices with a 26-month training dataset and eight-month ex situ validation dataset. A subsequent demonstration refines the approach, and is demonstrated across a panel of six high volume devices with a 29-month training dataset and a 13-month ex situ validation dataset. In both implementations, significant improvement is realised versus legacy forecasting systems.

AB - A model-based approach to the forecasting of short-range product demand within the semiconductor industry is presented. Device-level forecast models are developed via a novel two-stage stochastic algorithm that permits leading indicators to be optimally blended with smoothed estimates of unit-level demand. Leading indicators include backlog, bookings, delinquencies, inventory positions, and distributor resales. Group level forecasts are easily obtained through upwards aggregation of the device level forecasts. The forecasting algorithm is demonstrated at two major US-based semiconductor manufacturers. The first application involves a product family consisting of 254 individual devices with a 26-month training dataset and eight-month ex situ validation dataset. A subsequent demonstration refines the approach, and is demonstrated across a panel of six high volume devices with a 29-month training dataset and a 13-month ex situ validation dataset. In both implementations, significant improvement is realised versus legacy forecasting systems.

KW - Demand forecasting

KW - Enterprise resource planning

KW - Operations modelling

KW - Semiconductor fabrication

KW - Statistical forecasting

KW - Supply chain management

UR - http://www.scopus.com/inward/record.url?scp=41749121495&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=41749121495&partnerID=8YFLogxK

U2 - 10.1504/EJIE.2008.017686

DO - 10.1504/EJIE.2008.017686

M3 - Article

AN - SCOPUS:41749121495

VL - 2

SP - 253

EP - 278

JO - European Journal of Industrial Engineering

JF - European Journal of Industrial Engineering

SN - 1751-5254

IS - 3

ER -