Example-based know-how recycling, while applied easily to well-defined problems with a solid parameter set and lots of examples, can also enhance many solutions, for example in product design and business management, if the basis of examples is rather 'sparse', the list of relevant influence parameters is unreliable, and explicit 'scientific' know-how is not available. As an application example from the area of product design, refinement of the side door handling (SDH) comfort of cars is discussed: From a list of 78 potential infiuence parameters a 'best fitting' subset is selected with the aid of a step-wise multivariate non-linear regression procedure, driven by a genetic algorithm, which employs a family of neural networks as a rating model. Based on the parameter subsets found, neural models are constructed, which represent, for example the overall global SDH comfort with respect to relevant influence parameters and the 29 examples given. Employing the neural models, systematical refinement of eg. overall global SDH comfort can be achieved either by a step-wise process based on local derivatives or by a genetic optimization routine, and 'novelty' of a solution with respect to the 29 examples given can be taken into account, to avoid for example inadmissible extrapolation. Following this line, the SDH comfort of a car design can be refined systematically under given restrictions, in terms of parameters and 'novelty'. Accordingly, solutions for many similar 'sparse' and highly interdisciplinary problems can be supported in the same way, especially if no explicit know-how from experts or 'scientific modeling' is available.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Beispielbasiertes Know-how Recycling auch bei heterogenen Aufgabenstellungen mit 'Schütterer' Wissensgrundlage: Verbesserung des Seitentür-Betätigungskomforts von PKW


    Additional title:

    Example-based know-how recycling applied also to heterogeneous tasks with a sparse knowledge background: improving the handling comfort of car doors


    Contributors:
    Stricker, R. (author) / Huber, B. (author)


    Publication date :

    1998


    Size :

    11 Seiten, 10 Bilder, 5 Quellen




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    German





    SEITENTÜR-BEFESTIGUNGSSTRUKTUR

    TAKAHASHI SUGURU | European Patent Office | 2019

    Free access

    Kraftfahrzeug-Seitentür

    HINTENNACH MARKUS / SEITZ FRANZ / LEKAR JAN | European Patent Office | 2023

    Free access

    Seitentür eines Kraftfahrzeugs

    HOFMANN DOMINIK / MALHERBE LAURENT / RITTHALER EIKE et al. | European Patent Office | 2022

    Free access

    Seitentür für ein Fahrzeug sowie Fahrzeug mit einer derartigen Seitentür

    RODERMUND WILFRIED | European Patent Office | 2016

    Free access