IDENTIFICATION AND INTERPRETATION OF MANUFACTURING PROCESS PATTERNS THROUGH NEURAL NETWORKS

To produce products with consistent quality, manufacturing processes n eed to be closely monitored for any deviations in the process. Proper analysis of control charts that are used to determine the state of the process not only requires a thorough knowledge and understanding of t he underlying distribution theories associated with control charts, bu t also the experience of an expert in decision making. The present wor k proposes a modified backpropagation neural network methodology to id entify and interpret various patterns of variations that can occur in a manufacturing process. The neural network methodology is developed u tilizing the delta-bar-delta learning rule and hyperbolic tangent acti vation function. The methodology adopted is designed with the objectiv e to recognize both small and large magnitude deviations and also to i dentify the nature of process change, so that proper corrective action may be taken to remedy the problem. The network can identify patterns of variation such as shift patterns and trend patterns, as well as no rmal patterns with a fewer number of subgroups when compared with cont rol charts. This can be utilized to signal an out-of-control condition well in advance to facilitate the prevention of nonconforming product s from being produced.