Page 112 - MSAM-4-3

Materials Science in Additive Manufacturing Sustainable manufacturing composite material optimization

In addition, AI-powered sustainability frameworks are strength, flexural strength, impact resistance, and surface
being created to minimize the carbon footprint of FDM roughness, all of which are key attributes for impeller
printing. These frameworks support the alignment of performance in high-speed fluid applications.
AI-enabled manufacturing with the circular economy. Beyond mechanical performance, energy efficiency has
With AI capabilities under development, its potential in emerged as a pivotal consideration in FDM-based impeller
FDM will shift from parameter optimization to end-to- manufacturing. The integration of power meters into FDM
end process automation, where intelligent autonomous machines has enabled real-time monitoring of actual
systems automate the entire AM process, thereby enabling energy consumption, providing valuable data for process
a new paradigm of green, high-performance, and fully optimization. Empirical studies have demonstrated that
autonomous additive manufacturing. higher extrusion temperatures enhance interlayer adhesion

1.2. MCDM in FDM process optimization in impeller but significantly increase power consumption, while
production lower extrusion temperatures reduce energy usage at the
expense of weakened interlayer bonding. 37-39 In response
The application of MCDM methods has gained significant to these trade-offs, researchers have developed AI-assisted
attention in optimizing the FDM process for fabricating MCDM frameworks that incorporate real-time energy
impellers, likely due to challenges in identifying an data into the decision-making process. These intelligent
optimal set of process parameters that balance mechanical systems enable dynamic optimization of parameters such
performance, energy efficacy, and material utilization. as extrusion temperature, print speed, and cooling settings,
Impellers, which are critical components in pumps, aiming to minimize power consumption while preserving
turbines, and compressors, demand high precision in mechanical integrity.
terms of geometric accuracy, structural integrity, and
fluid flow dynamics. 28-30 The most significant concern in Surface roughness is another critical performance
the FDM-based production of impellers is the intrinsic determinant for impellers, as it directly affects fluid
trade-offs amongst conflicting objectives, such as efficiency and resistance to cavitation. Elevated surface
maximizing mechanical strength and surface finish while roughness promotes turbulent flow and energy losses,
simultaneously minimizing printing time and energy ultimately diminishing the impeller’s overall efficiency.
utilization. Conventional optimization methods, such as Conventionally, improving surface finish requires post-
trial-and-error and single-objective optimization models, processing techniques such as sanding or chemical
proved inadequate in portraying these multi-dimensional smoothing – methods that are both time-consuming
interactions, warranting the implementation of advanced and costly. Recent MCDM-based studies, however, have
MCDM paradigms. Recent studies have demonstrated the focused on in-process optimization by tuning parameters
effectiveness of hybrid MCDM methods that combine the such as layer height, nozzle temperature, and print
application of Fuzzy AHP, TOPSIS, and genetic algorithms speed, which naturally influence surface quality. Hybrid
(GA) to systematically select and rank the best process optimization frameworks that integrate GA with MCDM
parameters for FDM-fabricated impellers. 31-33 approaches have demonstrated success in automating the
selection of process parameters to achieve low surface
Mechanical integrity is one of the most critical
performance parameters for impellers, and it is influenced roughness. These methods reduce or eliminate the need
for post-processing, thereby enhancing manufacturing
by a dynamic set of process parameters, including efficiency and cost-effectiveness in FDM-based impeller
extrusion temperature, layer thickness, infill density, and production. 40-42
print speed. Prior research has demonstrated that higher
infill density and lower layer thickness can significantly The integration of MCDM methods with AI has further
improve mechanical strength and surface finish, but these advanced impeller optimization by enabling predictive
adjustments typically result in higher energy consumption modeling and real-time process control. Leveraging ML
and a longer print time. Conversely, a higher print speed algorithms trained on historical print data, researchers
34
reduces production time but compromises interlayer have developed adaptive MCDM models capable of
adhesion and structural integrity. To avoid such trade- predicting defects and dynamically adjusting process
35
offs, Fuzzy AHP has been utilized to assign weights to parameters during fabrication. This approach has led
42
performance criteria based on expert evaluations. These to significantly higher first-print success rates, thereby
weights are then incorporated into TOPSIS to rank various minimizing material waste and improving overall
parameter sets in terms of closeness to an ideal solution. 35,36 production efficiency. In addition, the incorporation of
This hybrid approach enables the selection of optimal metaheuristic optimization techniques, such as the Non-
process parameters that simultaneously maximize tensile dominated Sorting Genetic Algorithm II (NSGA-II) and

Volume 4 Issue 3 (2025) 4 doi: 10.36922/MSAM025200033