Please use this identifier to cite or link to this item: doi:10.22028/D291-38945
Title: Foundation of expert systems for conceptional design in mechanical engineering
Author(s): Raulefs, Peter
Language: English
Year of Publication: 1984
Place of publication: Kaiserslautern
DDC notations: 004 Computer science, internet
Publikation type: Report
Abstract: A theory of qualitative reasoning on physico-technical systems is developed. Particular results include • an approach to reasoning on time and causality combining state- and event-based approaches allowing concurrent time structures. • the multiple view technique which allows to represent, reason about, and relate different views of time scales and other physical dimensions with respect to e.g. precision and granularity. Precise quantitative models appear as particular views, so that conventional mathematical reasoning can be incorporated in our approach. • representing physico-technical systems in terms of objects with internal states, processes denoting consecutive and/or concurrent structures of state transitions (events), and meta-processes for expressing teleological considerations. This approach is summarized in terms of the qualitative reasoning language QRL which, in addition, provides various aggregation and abstraction mechanisms on objects and processes. • an approach for applying non-monotonic reasoning techniques to the design of physico-technical systems. These results provide a foundation towards mechanizing the construction process in mechanical engineering: QRL supports the phases of requirements definition and specification, as well as conceptional design in a way which is refinable to material design as supported by CAD-systems.
Link to this record: urn:nbn:de:bsz:291--ds-389456
hdl:20.500.11880/35461
http://dx.doi.org/10.22028/D291-38945
Series name: Memo SEKI : SEKI-Projekt / Deutsches Forschungszentrum für Künstliche Intelligenz, DFKI
Series volume: 84,8
Date of registration: 20-Mar-2023
Faculty: SE - Sonstige Einrichtungen
Department: SE - DFKI Deutsches Forschungszentrum für Künstliche Intelligenz
Professorship: SE - Sonstige
Collections:SciDok - Der Wissenschaftsserver der Universität des Saarlandes



Items in SciDok are protected by copyright, with all rights reserved, unless otherwise indicated.