Designing Robust, cost-effective field measurement sets using universal multiple linear regression

Melissa Clutter, Paul A Ferre

Research output: Contribution to journalArticle

Abstract

Limited monitoring budgets restrict the type and number of sensors that can be installed for field-based studies. Therefore, sensor selection should be both informative and efficient. We propose a method to optimize sensor network design, prior to data collection, by combining multiple linear regression (MLR) and robust decision-making (RDM). Multiple linear regression inherently considers the strength of the relationship between observations and predictions of interest and correlations among proposed observations. In our approach, we use universal Multiple Linear Regression (uMLR) to quantify the explanatory power of all possible combinations of model-simulated candidate observations (of different sensor types and locations). A modelensemble approach allows for network design in the context of user-defined uncertainties, including expected measurement error and parameter and structural uncertainty. Application of uMLR with RDM produces a comprehensive assessment of the likely value of many observation sets. These results can be used to design sensor networks to address specific experimental objectives and to balance the cost and effort of installing sensors to the expected value of the data for model testing and decision support.

Original languageEnglish (US)
Pages (from-to)531-541
Number of pages11
JournalSoil Science Society of America Journal
Volume83
Issue number3
DOIs
StatePublished - Jan 1 2019

Fingerprint

sensors (equipment)
sensor
cost
network design
decision making
uncertainty
prediction
monitoring
testing

ASJC Scopus subject areas

  • Soil Science

Cite this

Designing Robust, cost-effective field measurement sets using universal multiple linear regression. / Clutter, Melissa; Ferre, Paul A.

In: Soil Science Society of America Journal, Vol. 83, No. 3, 01.01.2019, p. 531-541.

Research output: Contribution to journalArticle

@article{8d15c17a80af458591b2bd6cb5dfe315,
title = "Designing Robust, cost-effective field measurement sets using universal multiple linear regression",
abstract = "Limited monitoring budgets restrict the type and number of sensors that can be installed for field-based studies. Therefore, sensor selection should be both informative and efficient. We propose a method to optimize sensor network design, prior to data collection, by combining multiple linear regression (MLR) and robust decision-making (RDM). Multiple linear regression inherently considers the strength of the relationship between observations and predictions of interest and correlations among proposed observations. In our approach, we use universal Multiple Linear Regression (uMLR) to quantify the explanatory power of all possible combinations of model-simulated candidate observations (of different sensor types and locations). A modelensemble approach allows for network design in the context of user-defined uncertainties, including expected measurement error and parameter and structural uncertainty. Application of uMLR with RDM produces a comprehensive assessment of the likely value of many observation sets. These results can be used to design sensor networks to address specific experimental objectives and to balance the cost and effort of installing sensors to the expected value of the data for model testing and decision support.",
author = "Melissa Clutter and Ferre, {Paul A}",
year = "2019",
month = "1",
day = "1",
doi = "10.2136/sssaj2018.09.0340",
language = "English (US)",
volume = "83",
pages = "531--541",
journal = "Soil Science Society of America Journal",
issn = "0361-5995",
publisher = "Soil Science Society of America",
number = "3",

}

TY - JOUR

T1 - Designing Robust, cost-effective field measurement sets using universal multiple linear regression

AU - Clutter, Melissa

AU - Ferre, Paul A

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Limited monitoring budgets restrict the type and number of sensors that can be installed for field-based studies. Therefore, sensor selection should be both informative and efficient. We propose a method to optimize sensor network design, prior to data collection, by combining multiple linear regression (MLR) and robust decision-making (RDM). Multiple linear regression inherently considers the strength of the relationship between observations and predictions of interest and correlations among proposed observations. In our approach, we use universal Multiple Linear Regression (uMLR) to quantify the explanatory power of all possible combinations of model-simulated candidate observations (of different sensor types and locations). A modelensemble approach allows for network design in the context of user-defined uncertainties, including expected measurement error and parameter and structural uncertainty. Application of uMLR with RDM produces a comprehensive assessment of the likely value of many observation sets. These results can be used to design sensor networks to address specific experimental objectives and to balance the cost and effort of installing sensors to the expected value of the data for model testing and decision support.

AB - Limited monitoring budgets restrict the type and number of sensors that can be installed for field-based studies. Therefore, sensor selection should be both informative and efficient. We propose a method to optimize sensor network design, prior to data collection, by combining multiple linear regression (MLR) and robust decision-making (RDM). Multiple linear regression inherently considers the strength of the relationship between observations and predictions of interest and correlations among proposed observations. In our approach, we use universal Multiple Linear Regression (uMLR) to quantify the explanatory power of all possible combinations of model-simulated candidate observations (of different sensor types and locations). A modelensemble approach allows for network design in the context of user-defined uncertainties, including expected measurement error and parameter and structural uncertainty. Application of uMLR with RDM produces a comprehensive assessment of the likely value of many observation sets. These results can be used to design sensor networks to address specific experimental objectives and to balance the cost and effort of installing sensors to the expected value of the data for model testing and decision support.

UR - http://www.scopus.com/inward/record.url?scp=85068691999&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068691999&partnerID=8YFLogxK

U2 - 10.2136/sssaj2018.09.0340

DO - 10.2136/sssaj2018.09.0340

M3 - Article

AN - SCOPUS:85068691999

VL - 83

SP - 531

EP - 541

JO - Soil Science Society of America Journal

JF - Soil Science Society of America Journal

SN - 0361-5995

IS - 3

ER -