Robustness of recovery in locating array-based screening experiments

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Locating arrays (LAs) are experimental designs for screening interactions in engineered systems. LAs are often highly unbalanced, requiring advanced techniques to recover the terms that significantly influence system performance. While perfect recovery is achieved in the absence of noise, real systems are noisy. Therefore, in this paper, we study the robustness of recovery in the presence of noise. Using known models to generate synthetic data, we investigate recovery accuracy as a function of noise. Separation is introduced into LAs to allow more coverage for each t-way interaction; when separation is higher, recovery in noisy scenarios should improve. We find that locating arrays are able to recover the influential terms even with high levels of noise and that separation appears to improve recovery. Under the pessimistic assumption that noise depends on the range of responses, it is no surprise that terms with small coefficients become indistinguishable from noise.

Original languageEnglish (US)
Title of host publicationSimulation Series
PublisherThe Society for Modeling and Simulation International
Edition4
ISBN (Electronic)9781510892521, 9781510892538, 9781510892545, 9781510892552, 9781510892569
DOIs
StatePublished - Jan 1 2019
Event2019 Communications and Networking Simulation, CNS 2019, Part of the 2019 Spring Simulation Multi-Conference, SpringSim 2019 - Tucson, United States
Duration: Apr 29 2019May 2 2019

Publication series

NameSimulation Series
Number4
Volume51
ISSN (Print)0735-9276

Conference

Conference2019 Communications and Networking Simulation, CNS 2019, Part of the 2019 Spring Simulation Multi-Conference, SpringSim 2019
CountryUnited States
CityTucson
Period4/29/195/2/19

Fingerprint

Screening
Recovery
Experiments
Design of experiments

Keywords

  • Analysis
  • Locating arrays
  • Recovery
  • Robustness
  • Screening engineered networks

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Seidel, S. A., Colbourn, C. J., & Syrotiuk, V. R. (2019). Robustness of recovery in locating array-based screening experiments. In Simulation Series (4 ed.). (Simulation Series; Vol. 51, No. 4). The Society for Modeling and Simulation International. https://doi.org/10.23919/SpringSim.2019.8732911

Robustness of recovery in locating array-based screening experiments. / Seidel, Stephen A.; Colbourn, Charles J.; Syrotiuk, Violet R.

Simulation Series. 4. ed. The Society for Modeling and Simulation International, 2019. (Simulation Series; Vol. 51, No. 4).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Seidel, SA, Colbourn, CJ & Syrotiuk, VR 2019, Robustness of recovery in locating array-based screening experiments. in Simulation Series. 4 edn, Simulation Series, no. 4, vol. 51, The Society for Modeling and Simulation International, 2019 Communications and Networking Simulation, CNS 2019, Part of the 2019 Spring Simulation Multi-Conference, SpringSim 2019, Tucson, United States, 4/29/19. https://doi.org/10.23919/SpringSim.2019.8732911
Seidel SA, Colbourn CJ, Syrotiuk VR. Robustness of recovery in locating array-based screening experiments. In Simulation Series. 4 ed. The Society for Modeling and Simulation International. 2019. (Simulation Series; 4). https://doi.org/10.23919/SpringSim.2019.8732911
Seidel, Stephen A. ; Colbourn, Charles J. ; Syrotiuk, Violet R. / Robustness of recovery in locating array-based screening experiments. Simulation Series. 4. ed. The Society for Modeling and Simulation International, 2019. (Simulation Series; 4).
@inproceedings{853e5afafc464729bfadcf40368b8c57,
title = "Robustness of recovery in locating array-based screening experiments",
abstract = "Locating arrays (LAs) are experimental designs for screening interactions in engineered systems. LAs are often highly unbalanced, requiring advanced techniques to recover the terms that significantly influence system performance. While perfect recovery is achieved in the absence of noise, real systems are noisy. Therefore, in this paper, we study the robustness of recovery in the presence of noise. Using known models to generate synthetic data, we investigate recovery accuracy as a function of noise. Separation is introduced into LAs to allow more coverage for each t-way interaction; when separation is higher, recovery in noisy scenarios should improve. We find that locating arrays are able to recover the influential terms even with high levels of noise and that separation appears to improve recovery. Under the pessimistic assumption that noise depends on the range of responses, it is no surprise that terms with small coefficients become indistinguishable from noise.",
keywords = "Analysis, Locating arrays, Recovery, Robustness, Screening engineered networks",
author = "Seidel, {Stephen A.} and Colbourn, {Charles J.} and Syrotiuk, {Violet R.}",
year = "2019",
month = "1",
day = "1",
doi = "10.23919/SpringSim.2019.8732911",
language = "English (US)",
series = "Simulation Series",
publisher = "The Society for Modeling and Simulation International",
number = "4",
booktitle = "Simulation Series",
edition = "4",

}

TY - GEN

T1 - Robustness of recovery in locating array-based screening experiments

AU - Seidel, Stephen A.

AU - Colbourn, Charles J.

AU - Syrotiuk, Violet R.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Locating arrays (LAs) are experimental designs for screening interactions in engineered systems. LAs are often highly unbalanced, requiring advanced techniques to recover the terms that significantly influence system performance. While perfect recovery is achieved in the absence of noise, real systems are noisy. Therefore, in this paper, we study the robustness of recovery in the presence of noise. Using known models to generate synthetic data, we investigate recovery accuracy as a function of noise. Separation is introduced into LAs to allow more coverage for each t-way interaction; when separation is higher, recovery in noisy scenarios should improve. We find that locating arrays are able to recover the influential terms even with high levels of noise and that separation appears to improve recovery. Under the pessimistic assumption that noise depends on the range of responses, it is no surprise that terms with small coefficients become indistinguishable from noise.

AB - Locating arrays (LAs) are experimental designs for screening interactions in engineered systems. LAs are often highly unbalanced, requiring advanced techniques to recover the terms that significantly influence system performance. While perfect recovery is achieved in the absence of noise, real systems are noisy. Therefore, in this paper, we study the robustness of recovery in the presence of noise. Using known models to generate synthetic data, we investigate recovery accuracy as a function of noise. Separation is introduced into LAs to allow more coverage for each t-way interaction; when separation is higher, recovery in noisy scenarios should improve. We find that locating arrays are able to recover the influential terms even with high levels of noise and that separation appears to improve recovery. Under the pessimistic assumption that noise depends on the range of responses, it is no surprise that terms with small coefficients become indistinguishable from noise.

KW - Analysis

KW - Locating arrays

KW - Recovery

KW - Robustness

KW - Screening engineered networks

UR - http://www.scopus.com/inward/record.url?scp=85073698024&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073698024&partnerID=8YFLogxK

U2 - 10.23919/SpringSim.2019.8732911

DO - 10.23919/SpringSim.2019.8732911

M3 - Conference contribution

AN - SCOPUS:85073698024

T3 - Simulation Series

BT - Simulation Series

PB - The Society for Modeling and Simulation International

ER -