The methodology of predictive design analysis

Martin Eriksson

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

Abstract

From an engineering point of view, uncertainty is present in all areas of design (products, processes and organizations). Computer-based design analysis, here confined to quantitative computer-based structural design analysis within mechanical engineering, serves as an important source of information in decisions taken during the design activity; importantly, it aims at giving confidence in critical design results. It is moreover nowadays used on all levels of concretization of the product-to-be throughout the development process. It must therefore address different uncertainties and errors during the whole development process. To that end, the concept of predictive design analysis (PDA) was introduced. The initial version of PDA treated primarily uncertainties of aleatory nature and was confined to product/technology-related issues. Today a broader perspective on the uncertainties is needed, and it is important to develop reliable design analysis methods because of the increasing use of design analysis by both analysts and engineering designers. Hence it is therefore necessary to extend the PDA framework into a full-blown methodology. PDA is a specific computer-based design analysis methodology that supports the systematic handling of uncertainties and errors during the computer-based design analysis activity throughout the whole development of the artifact. Such a methodology includes: Not only aleatory uncertainties, but also epistemic uncertainties connected with factors affecting the design analysis activities; Operationally efficient and effective integration between the engineering design and design analysis activities; Quality assurance aspects in terms of quality checks, verification and validation activities taking physical testing into account; Progress monitoring throughout all design analysis activities from clarification to completion; Traceability in utilized information, technologies and established results; Information and knowledge re-use for an improved uncertainty treatment in future design analysis activities through establishment of lessons learned and best practice documentation as well as methodology development.
In this paper the synthesis of the PDA methodology is reported. Starting from an elaboration on different concepts of uncertainty, the constituent elements of the extended PDA methodology are presented. The methodology is then illustrated with a case study.
Original languageEnglish
Title of host publication[Host publication title missing]
PublisherAmerican Society Of Mechanical Engineers (ASME)
Number of pages15
VolumeVolume 11: Systems, Design, and Complexity
ISBN (Print)978-0-7918-4960-6
DOIs
Publication statusPublished - 2014
EventInternational Mechanical Engineering Congress & Exposition - IMECE2014 - Montreal, Quebec, CANADA
Duration: 2014 Nov 142014 Nov 20

Publication series

Name
VolumeVolume 11: Systems, Design, and Complexity

Conference

ConferenceInternational Mechanical Engineering Congress & Exposition - IMECE2014
Period2014/11/142014/11/20

Subject classification (UKÄ)

  • Production Engineering, Human Work Science and Ergonomics

Free keywords

  • Maskinkonstruktion
  • Engineering Design
  • Machine Design
  • Uncertainty
  • Predictive Design Analysis

Fingerprint

Dive into the research topics of 'The methodology of predictive design analysis'. Together they form a unique fingerprint.

Cite this