Process diagnostics as a means of improving the efficiency of quality control

Abstract

The new requirements for products quality and manufacturing process productivity can not be met by using traditional methods. This paper presents a new approach to improving the efficiency of quality control in machining. The method is especially suitable for quality control with the use of Shewhart's control charts (SCC). It is based on processing signals that accompany machining processes (e.g. acoustic emission, force, vibrations) with the use of fuzzy reasoning and neural network. The on-line measured signals are transformed through the process safety function (PSF) which is developed on the basis of two quantities specially defined for this method: the process quality of performance (QOP) and the process state vector (PSV). It is finally used for estimation of the process safety index (PSI). Inference models developed in this work can easily be identified and modified under workshop circumstances.