Akademska digitalna zbirka SLovenije - logo
VSE knjižnice (vzajemna bibliografsko-kataložna baza podatkov COBIB.SI)
  • Qualitative assessment of machining technologies using inductive machine learning
    Junkar, Mihael ; Levy, Paul ; Filipič, Bogdan
    In the past much intensive research has been carried out in the field of machining process identification, modelling and simulation. Many of the results obtained have led to the implementation of ... process control and optimisationin the industrial environment. The traditional methods used were based on deterministic approaches using exhaustive "cut and try" methods, resulting in mathematical models which more or less diverge from the real process. On the other hand probabilistic methods were used resulting in modelsrepresented as transfer functions. Further on new approaches based on artificial intelligence, neural networks and chaos theory were applied. We can now already encounter quite a number of different types of machining equipmentwhich can be purchased in the marked place based on a synergetic approach, which combines and integrates these different approaches. This paper discusses the problems related to the "human-to-machine interface" whichin our view has to be taken into account as early as the design phase of the machining process device, its controlling technology and use if implementation is to be effective. Since only a part of information and knowledge of the process can be acquired by quantitative measurements, the introduction of knowledge acquisition based on qualitative measurements, (operator's interviews and observations) is proposed as the basis for classification and improvement of process performance. The possibility of the same approach used at the level of interfunctional communication facilitating the knowledge transfer from the shop floor level to the planning function enabling better decision making is also discussed. Overall the paper provides a discussion of the relationship between the qualitative aspects of the operator-technologyinterface andsuggests that such data is crucial for the effective development of Ai-based decision making and technology design. The paper presents empirical observations from organisations adopting a range of conventional and nonconventional machining applications.
    Vir: Manufacturing systems = Fertigungssysteme = Systemes de fabrication : proceedings of the CIRP seminars. - ISSN 0176-3377 (Vol. 25, no. 3, 1996, str. 219-224)
    Vrsta gradiva - članek, sestavni del
    Leto - 1996
    Jezik - angleški
    COBISS.SI-ID - 1991707

Vnos na polico