Decision-making in product quality based on failure knowledge

Wei Dai*, Paul G. Maropoulos, Wai Ming Cheung, Xiaoqing Tang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


Decision-making in product quality is an indispensable stage in product development, in order to reduce product development risk. Based on the identification of the deficiencies of quality function deployment (QFD) and failure modes and effects analysis (FMEA), a novel decision-making method is presented that draws upon a knowledge network of failure scenarios. An ontological expression of failure scenarios is presented together with a framework of failure knowledge network (FKN). According to the roles of quality characteristics (QCs) in failure processing, QCs are set into three categories namely perceptible QCs, restrictive QCs, and controllable QCs, which present the monitor targets, control targets and improvement targets respectively for quality management. A mathematical model and algorithms based on the analytic network process (ANP) is introduced for calculating the priority of QCs with respect to different development scenarios. A case study is provided according to the proposed decision-making procedure based on FKN. This methodology is applied in the propeller design process to solve the problem of prioritising QCs. This paper provides a practical approach for decision-making in product quality.

Original languageEnglish
Pages (from-to)143-163
Number of pages21
JournalInternational Journal of Product Lifecycle Management
Issue number2-4
Publication statusPublished - Oct 2011
Externally publishedYes


  • Analytic network process
  • ANP
  • Decision-making in product quality
  • Decision-making model
  • Failure knowledge network
  • FKN

ASJC Scopus subject areas

  • Business and International Management
  • Safety, Risk, Reliability and Quality
  • Management Science and Operations Research


Dive into the research topics of 'Decision-making in product quality based on failure knowledge'. Together they form a unique fingerprint.

Cite this