RISK FAILURE REDUCTION IN 3D PRINTER THROUGH SECUENTIAL USE OF DFMEA, FAULT TREE AND BAYESIAN NETWORKS

Abstract

The need to manufacture reliable devices, which include features that guarantee good performance and do not cause problems for the end user, is of paramount importance for manufacturers. To meet this objective, it is necessary to perform a thorough analysis of the devices to identify potential failure events in order to be able to take actions to reduce their risk of occurrence and increase the reliability and quality of the device. This research paper presents an effective tool for the detection from the design of possible failures in the devices, which allows actions to be taken for their correction in early stages. This analysis methodology combines several advanced fault analysis techniques, such as DFMEA, Fault Tree, and Bayesian networks, making the process of analyzing, detecting and correcting potential device failures more efficient. This methodology is applied to the analysis of a commercial 3D printer that uses fused filament deposition technology model Anet A8, making a preliminary filter using a DFMEA for subsequent analysis fault tree and Bayesian network managing to determine the probability of occurrence of 3D printing failures, this allows to take actions and establish priorities of corrective actions focused on reducing the risk of failure based on its probability of occurrence. Keywords: DFMEA, Fault tree, Bayesian Network, 3D printing, Fault probability