A Multi-Criteria Decision-Making Approach for Enhancing Mechanical Properties of FDM 3D-Printed Part

Abstract

The quest for optimizing 3D printing processes to meet industrial demands for improved mechanical properties and dimensional stability is an ongoing challenge. This study delves into the task of determining the optimal 3D printing parameter (material and layer height) and annealing parameter (annealing time and annealing temperature) combinations for FDM 3D-printed parts through a systematic and objective approach. By utilizing Multi-Criteria Decision-Making (MCDM) methods, specifically the Analytic Hierarchy Process (AHP) for criteria weighting and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for ranking, the study endeavors to address this challenge effectively. The primary aim of this research is to identify the most suitable 3D printing and annealing parameter combination that maximizes tensile strength and Young's modulus of elasticity while minimizing change in width, change in thickness, and change in length of the specimen after annealing. To achieve this, a comprehensive dataset stemming from the experimental analysis of three distinct materials (Polylactic acid - PLA, Polyethylene terephthalate glycol - PETG, carbon fiber reinforced PETG - PETGCF), three layer heights (0.1 mm, 0.2 mm, 0.3 mm), five annealing temperatures (60°C, 70°C, 80°C, 90°C, 100°C), and three annealing times (30 minutes, 60 minutes, 90 minutes) is employed. These criteria were determined based on the requirements of a specific industrial case study, highlighting their relevance and significance in real-world applications. The results show that the best combinations are the ones from PETGCF material with 0,1 mm layer height, with long annealing times (90 minutes) and low to mid annealing temperatures (60 - 70°C). The worst alternatives are the ones annealed at high temperatures (90 - 100°C), and with PETG material, as the dimensional change of this material are significant at high temperatures.