Innovative methods are explored for computer based quality control of parts manufacture. The approach is based on the use of analysis of variance techniques and regression modelling to formulate a description (model) of the product and process in terms of product quality. This model is then employed in a computer based on-line quality control strategy in which the aim is to optimise manufacturing quality by attempting to compensate for product deviations.
Available quality control software is based on methodologies developed in the 1920s . It is concerned with the presentation of data and off-line analysis rather than the decision making process. By integrating statistics and quality control strategies, computer based decisions can be formulated based on larger quantities of data than a human could assimilate. Such decisions can be made more accurately and in a fraction of the time taken by manual means.
A case study is presented which is concerned with the implementation of computer based on-line quality control of manufacture. This case study was successfully completed with a reduction in the first time reject rate from thirty percent to below four percent. The case study shows the generic nature of the method and also the diversity of application.