Premium
1.1.3 A Quantitative Methodology for the Optimal Selection of Design Margins in Complex Systems
Author(s) -
Snape Simon,
Sen Pratyush,
Whittle Steve
Publication year - 2004
Publication title -
incose international symposium
Language(s) - English
Resource type - Journals
ISSN - 2334-5837
DOI - 10.1002/j.2334-5837.2004.tb00474.x
Subject(s) - probabilistic design , context (archaeology) , probabilistic logic , selection (genetic algorithm) , computer science , engineering design process , set (abstract data type) , process (computing) , design methods , systems engineering , reliability engineering , engineering , machine learning , artificial intelligence , mechanical engineering , paleontology , biology , programming language , operating system
Design margins are the additional performance capability incorporated into a system to compensate for uncertainties. It is common place for uncertainties to be allowed for by introducing design margins based on experience. However, there is currently great interest within industry in applying robust and probabilistic methods, as the arbitrary selection of design margins is inappropriate in today's design ‐ to ‐ cost environment. With this in mind, this paper presents a novel quantitative methodology developed to estimate and select design margins in an optimal and robust manner. Particular attention is given to design margins in a performance context, i.e. design margins intended to improve the probability of a system/process performing to requirements. Furthermore the link between Design Margins and Critical Design Features (a sub‐set of top level product attributes / requirements that are agreed as being critical to project success) is explained and an approach to determine and manage both is introduced.