Abstract
Keywords - Cause and Effect Analysis, Neural Networks, Regression Analysis, Expert Systems, Learning from Examples.
Abstract - The cause and effect relationship is complex for many manufacturing processes. The ability to learn causal relationships from diagnostic examples is extremely useful. This learning ability can help not only to quantify the influence of causes on defects for existing components but also to set up new process, material and design parameters to manufacture new components. In this paper, we compare and contrast neural network based models to regression analysis models for analysing and quantifying cause and effect relationships for manufacturing processes. A neural network approach that can adapt and learn from past examples, uses the data to extract any pre-existing relationships between the input and output variables. However, a regression analysis may become advantageous if some knowledge of the input-output relationship is already known.
In many real situations very few good quality training examples are generally available. A few limitations of multi layer feed forward neural networks in presence of such noisy, limited and sparse training data sets have been discovered, and because of such limitations it has been found difficult to use this technique as a robust tool by end users in any other manufacturing industry to analyse cause and effect relationships. The similarities and dissimilarities between neural network models and regression analysis models have been explored, and the advantages and limitations of both the techniques for analysing and quantifying cause and effect relationships for manufacturing processes have been discovered.
A new algorithm - for which an international patent application has been filed is proposed. It has also been shown that this algorithm can overcome the limitations of both neural network and regression analysis techniques.