Decision Trees for Secure and Transparent Equipment Failure Prediction in Cloud-Connected Manufacturing

Abstract

Ensuring that equipment is reliable and secure is becoming more important as industrial processes rely more and more on devices linked to the cloud. It provides a new method for predicting when cloud-connected industrial equipment may break down. Fixing issues with data integrity and interpretability, the suggested method seeks to make the prediction model more secure while simultaneously making it more transparent. The Decision Trees (DT) method shows strong predicting skills using a dataset of occurrences of equipment breakdowns and performance histories. An integrated security layer that encrypts sensitive data further protects the prediction process. Utilizing visualization techniques that reveal the model’s decision-making process helps to achieve transparency. Shown experimentally is the efficacy of the method in anticipating equipment breakdowns within a safe and transparent framework. By providing a dependable and interpretable solution for proactive measures against equipment failures, this study adds to the progress of predictive maintenance techniques in cloud-connected manufacturing.