ICADA: Intelligent computer aided defect analysis for castings

Ransing, RS and Srinivasan, MN and Lewis, RW (1995) ICADA: Intelligent computer aided defect analysis for castings. In: Journal of Intelligent Manufacturing, 6 (1). pp. 29-40.

PDF ICADA_ntelligent_computer.pdf - Published Version Restricted to Registered users only Download (971kB) | Request a copy

Abstract

An intelligent computer aided defect analysis (ICADA) system, based on artificial intelligence techniques, has been developed to identify design, process or material parameters which could be responsible for the occurrence of defective castings in a manufacturing campaign. The data on defective castings for a particular time frame, which is an input to the ICADA system, has been analysed. It was observed that a large proportion, i.e. 50-80% of all the defective castings produced in a foundry, have two, three or four types of defects occurring above a threshold proportion, say 10%. Also, a large number of defect types are either not found at all or found in a very small proportion, with a threshold value below 2%. An important feature of the ICADA system is the recognition of this pattern in the analysis. Thirty casting defect types and a large number of causes numbering between 50 and 70 for each, as identified in the AFS analysis of casting defects-the standard reference source for a casting process-constituted the foundation for building the knowledge base. Scientific rationale underlying the formation of a defect during the casting process was identified and 38 metacauses were coded. Process, material and design parameters which contribute to the metacauses were systematically examined and 112 were identified as rootcauses. The interconnections between defects, metacauses and rootcauses were represented as a three tier structured graph and the handling of uncertainty in the occurrence of events such as defects, metacauses and rootcauses was achieved by Bayesian analysis. The hill climbing search technique, associated with forward reasoning, was employed to recognize one or several root causes.

Item Type:	Journal Article
Publication:	Journal of Intelligent Manufacturing
Publisher:	Springer
Additional Information:	Copyright of this article belongs to Springer.
Department/Centre:	Division of Mechanical Sciences > Mechanical Engineering
Date Deposited:	31 May 2011 06:40
Last Modified:	31 May 2011 06:40
URI:	http://eprints.iisc.ac.in/id/eprint/38028

Actions (login required)

View Item