ANFIS-based Models for Coating Quality Prediction for Thin-Film Deposition Processes

Thin-film deposition processes have gained much popularity due to their unique capability to enhance the physical and chemical properties of various materials. Identification of the best parametric combination for a deposition process to achieve desired coating quality is often considered to be challenging due to the involvement of a large number of input process parameters and conflicting responses. This study discusses the development of adaptive neuro-fuzzy inference system-based models for the prediction of quality measures of two thin-film deposition processes, i.e., SiCN thin-film coating using thermal chemical vapor deposition (CVD) process and Ni–Cr alloy thin-film coating using direct current magnetron sputtering process. The predicted response values obtained from the developed models are validated and compared based on actual experimental results which exhibit a very close match between both the values. The corresponding surface plots obtained from the developed models illustrate the effect of each process parameter on the considered responses. These plots will help the operator in selecting the best parametric mix to achieve enhanced coating quality. Also, analysis of variance results identifies the importance of each process parameter in the determination of response values. The proposed approach can be applied to various deposition processes for modeling and prediction of observed response values. It will also assist as an operator in selecting the best parametric mix for achieving desired response values.