Set-based design by simulation of usage scenario coverage

Bernard Yannou*, Pierre Alain Yvars, Chris Hoyle, Wei Chen

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

44 Scopus citations


While the marketing literature has advocated for decades that new products should be designed for intended and anticipated consumer usages, the engineering literature mostly proposes optimisation of product performances independent of specific users' skills, anticipated usage scenarios, and competing products in the market. In contrast to tedious market studies which assume an existing market experience for products and optimisation approaches based upon static product performances, we propose an adaptable approach to designing a product or product family: the set-based design by usage coverage simulation. It starts with generating a usage scenario space for a set of representative users. Next, considering a candidate set of products, one proceeds to the constraint satisfaction problem computations of feasible usage scenarios, assuming that physics-based models of performances are available. The comparison between the expected and feasible usage scenarios at the scale of a single user leads to Usage Coverage Indicators and finally to a preferred product which best covers the usage scenario space. At the level of a targeted consumer group, the approach provides a market share simulation for competing products or members of a scale-based product family. The design of a family of jigsaws thoroughly illustrates our approach.

Original languageEnglish (US)
Pages (from-to)575-603
Number of pages29
JournalJournal of Engineering Design
Issue number8
StatePublished - Aug 2013


  • constraint satisfaction problem
  • set-based design
  • simulation under uncertainty
  • usage context-based design
  • usage coverage metrics

ASJC Scopus subject areas

  • Engineering(all)


Dive into the research topics of 'Set-based design by simulation of usage scenario coverage'. Together they form a unique fingerprint.

Cite this