Robust design requirements specification: a quantitative method for requirements development using quality loss functions

Søren Nygaard Pedersen, Martin Ebro Christensen, Thomas J. Howard

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Product requirements serve many purposes in the product development process. Most importantly, they are meant to capture and facilitate product goals and acceptance criteria, as defined by stakeholders. Accurately communicating stakeholder goals and acceptance criteria can be challenging and more often than not, requirements will be subject to simplification, causing ambiguity and uncertainty, with negative consequences for the company and the users. To prevent such incidences, a new approach for creating more complete requirements is proposed in this article. Grounded in robust design theory, the approach uses quality loss functions as one of the five principles, to visualise a more complete set of requirement information in a single figure. In order to validate the potential and applicability of the proposed approach, a new indicator for requirement completeness is introduced, expressing how open the requirements are for interpretation. By applying the method and indicator to a case study from the medical device industry, it was found that less than 45% of the potential for quantification had been utilised. Finally, the robust design requirements specification method was successfully applied to three case study requirements, to illustrate the gains in terms of the level of quantification, transparency, and comprehensiveness of the provided information.
Original languageEnglish
JournalJournal of Engineering Design
Volume27
Issue number8
Pages (from-to)544-567
Number of pages24
ISSN0954-4828
DOIs
Publication statusPublished - 2016

Keywords

  • Customer satisfaction
  • Quality issues
  • Risk and reliability
  • Taguchi methods

Cite this

@article{e8a02016fb5749e38c14d681634791e7,
title = "Robust design requirements specification: a quantitative method for requirements development using quality loss functions",
abstract = "Product requirements serve many purposes in the product development process. Most importantly, they are meant to capture and facilitate product goals and acceptance criteria, as defined by stakeholders. Accurately communicating stakeholder goals and acceptance criteria can be challenging and more often than not, requirements will be subject to simplification, causing ambiguity and uncertainty, with negative consequences for the company and the users. To prevent such incidences, a new approach for creating more complete requirements is proposed in this article. Grounded in robust design theory, the approach uses quality loss functions as one of the five principles, to visualise a more complete set of requirement information in a single figure. In order to validate the potential and applicability of the proposed approach, a new indicator for requirement completeness is introduced, expressing how open the requirements are for interpretation. By applying the method and indicator to a case study from the medical device industry, it was found that less than 45{\%} of the potential for quantification had been utilised. Finally, the robust design requirements specification method was successfully applied to three case study requirements, to illustrate the gains in terms of the level of quantification, transparency, and comprehensiveness of the provided information.",
keywords = "Customer satisfaction, Quality issues, Risk and reliability, Taguchi methods",
author = "Pedersen, {S{\o}ren Nygaard} and Christensen, {Martin Ebro} and Howard, {Thomas J.}",
year = "2016",
doi = "10.1080/09544828.2016.1183163",
language = "English",
volume = "27",
pages = "544--567",
journal = "Journal of Engineering Design",
issn = "0954-4828",
publisher = "CRC Press/Balkema",
number = "8",

}

Robust design requirements specification: a quantitative method for requirements development using quality loss functions. / Pedersen, Søren Nygaard; Christensen, Martin Ebro; Howard, Thomas J.

In: Journal of Engineering Design, Vol. 27, No. 8, 2016, p. 544-567.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Robust design requirements specification: a quantitative method for requirements development using quality loss functions

AU - Pedersen, Søren Nygaard

AU - Christensen, Martin Ebro

AU - Howard, Thomas J.

PY - 2016

Y1 - 2016

N2 - Product requirements serve many purposes in the product development process. Most importantly, they are meant to capture and facilitate product goals and acceptance criteria, as defined by stakeholders. Accurately communicating stakeholder goals and acceptance criteria can be challenging and more often than not, requirements will be subject to simplification, causing ambiguity and uncertainty, with negative consequences for the company and the users. To prevent such incidences, a new approach for creating more complete requirements is proposed in this article. Grounded in robust design theory, the approach uses quality loss functions as one of the five principles, to visualise a more complete set of requirement information in a single figure. In order to validate the potential and applicability of the proposed approach, a new indicator for requirement completeness is introduced, expressing how open the requirements are for interpretation. By applying the method and indicator to a case study from the medical device industry, it was found that less than 45% of the potential for quantification had been utilised. Finally, the robust design requirements specification method was successfully applied to three case study requirements, to illustrate the gains in terms of the level of quantification, transparency, and comprehensiveness of the provided information.

AB - Product requirements serve many purposes in the product development process. Most importantly, they are meant to capture and facilitate product goals and acceptance criteria, as defined by stakeholders. Accurately communicating stakeholder goals and acceptance criteria can be challenging and more often than not, requirements will be subject to simplification, causing ambiguity and uncertainty, with negative consequences for the company and the users. To prevent such incidences, a new approach for creating more complete requirements is proposed in this article. Grounded in robust design theory, the approach uses quality loss functions as one of the five principles, to visualise a more complete set of requirement information in a single figure. In order to validate the potential and applicability of the proposed approach, a new indicator for requirement completeness is introduced, expressing how open the requirements are for interpretation. By applying the method and indicator to a case study from the medical device industry, it was found that less than 45% of the potential for quantification had been utilised. Finally, the robust design requirements specification method was successfully applied to three case study requirements, to illustrate the gains in terms of the level of quantification, transparency, and comprehensiveness of the provided information.

KW - Customer satisfaction

KW - Quality issues

KW - Risk and reliability

KW - Taguchi methods

U2 - 10.1080/09544828.2016.1183163

DO - 10.1080/09544828.2016.1183163

M3 - Journal article

VL - 27

SP - 544

EP - 567

JO - Journal of Engineering Design

JF - Journal of Engineering Design

SN - 0954-4828

IS - 8

ER -