Manufacturing Quality Prediction Model and Optimization of Operational Conditions for the Intelligentization of an Electroplating Process

Purpose: This study aims to develop an artificial intelligence (AI) model that predicts product quality by considering the interaction of various process parameters and controlling them to ensure the optimal combination of process variables.BRMethods: We developed a production quality prediction model using machine learning and particle swarm optimization (PSO). Based on this model, we explored optimal process conditions that maximize the likelihood of producing high-quality products.BRResults: By evaluating the quality prediction (classification) performance of various machine learning candidate models, we achieved an accuracy of over 95%. Using this model, we formulated an optimization problem to identify optimal process conditions.BRConclusion: The selected production quality prediction model was based on a support vector machine (SVM) model enhanced with the synthetic minority over-sampling technique (SMOTE). A process control chart was developed using the PSO algorithm to identify optimal process conditions. This approach enables the intelligent control of the electroplating process, thus improving production quality.