Analysing the Quality of Die Cast Components Using Expert System Related Technologies

An intelligent tool has been developed to analyse the quality of diecast components. The influence of both design and process parameters has been studied in great detail. The tool, based on expert systems related technologies, can identify the causes for the defective castings manufactured. Design related causes such as gate position; gate area, runner design etc. as well as process related causes such as cycle time, cavity filling time and melt temperature are suggested by analysing the data on defective castings for a manufacturing campaign. This intelligent analysis is useful for both design as well as production engineers in the decision making process. All the causes for the occurrence of defects are classified into two categories. The firs category is the 'rootcauses' i.e. the causes which are related to the actual process, material and design parameters and which can be directly monitored. The second category is more abstract in nature and is classified as `metacauses'. These capture the scientific rationale implicit in the cause and defect relationships. This knowledge on causes and defects is represented in the form of a structured graph rather than by a set of rules, which, coupled with forward reasoning and heuristic search techniques results in an elegant and efficient approach. Uncertainties and the degree of relationship among causes and defects are represented by a fuzzy probabilistic model.