A knowledge-based advisor for the automatic selection and sequencing of orienting devices for vibratory feeding

✍ Scribed by P.S. Tan; B.K.A. Ngoi; S.S.G. Lee; L.E.N. Lim

• Publisher: Elsevier Science
• Year: 1995
• Tongue: English
• Weight: 842 KB
• Volume: 8
• Category: Article
• ISSN: 0952-1976
• DOI: 10.1016/0952-1976(94)00053-p

✦ Synopsis

The selection and sequencing of orienting devices used in vibratory feeders is integral to the design of the feeders. However, because of the lack of organized, published guidelines, the selection and sequencing of orienting devices depends almost entirely on the experience of the skilled artisans who tool up the vibratory feeders. This paper discusses a knowledge-based framework by which a CAD model of the part to be oriented and fed is analysed for its significant orienting feature(s), and a suitable match of orienting devices is determined. The outcome of a consultation with the expert advisor typically yields more than one feasible sequence of orienting devices, leaving the user free to decide on the most appropriate sequence to adopt. The paper also discusses the hierarchical classification of the features of the part, as well as those of the orienting devices. The knowledge-based advisor was implemented on a 486-DX33 personal computer within the Microsoft ~ Windows 3.1 environment, which integrates both the CAD and the expert system. A benchmarking exercise carried out for two rotational and two prismatic parts yielded solutions that compared favorably with those used in industry, and those proposed by the UMASS system.