Return to search

Materials selection using knowledge-based techniques

A successful design is one of the most important elements for the commercial success of a product and the selection of appropriate materials is a key step within the product design process. The task is not easy; a large number of interacting factors, both technical and economic, need to be taken into consideration and a vast amount of data investigated. Product designers can benefit from using computer systems which can emulate the reasoning processes of an expert in selecting materials and provide ready access to appropriate materials data. The knowledge based system developed, Plassel fulfils the key requirements identified for such a system. It can: 1. Emulate the reasoning processes of a plastics expert. 2. Allow a customised data search to be undertaken 3. Access a range of data sources covering both embodiment and detail data. 4. Convert component functional requirements into property requirements. 5. Allow knowledge and experience to be stored in the system 6. Allow cost to be fully considered Professor Ashby in 1993 [1] stated "A full expert system for materials selection is decades away. Success has been achieved in specialised highly focused applications". Plasse1 is not such an application, it provides access to a full set of selection facilities. Novel aspects of Plassel include its ability to select on multi-dimensional criteria, automatically 'rate' materials and to conduct customised searches. Professor Ashby concludes with "It is only a question of time before more fully developed systems become available. They are something to keep informed about." Plassel is a more fully developed system for plastic materials selection than those currently available.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:319764
Date January 1995
CreatorsBal, Darbara Jay
PublisherUniversity of Warwick
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://wrap.warwick.ac.uk/80317/

Page generated in 0.0018 seconds