Page 114 - MSAM-4-3

Materials Science in Additive Manufacturing Sustainable manufacturing composite material optimization

efficiency. By analyzing historical energy usage, print job 2. Methods
complexity, and machine utilization patterns, AI algorithms
can schedule prints during off-peak hours, capitalizing 2.1. Experimental workflow for sustainable FDM
on cheaper and cleaner energy sources. In industrial AM printing of thermoplastic polyurethane (TPU) 95A
settings, multi-objective AI optimization frameworks are components
increasingly employed to simultaneously reduce material In this study, we focused on the fabrication, optimization,
consumption, energy demand, and build time, ensuring and mechanical characterization of TPU 95A specimens
both environmental and economic viability. AI-integrated produced through FDM, with particular focus on analyzing
47
MCDM tools, such as Fuzzy AHP and TOPSIS, enable energy consumption throughout each stage of the process.
manufacturers to make informed decisions based on trade- The objective is to assess the efficiency, mechanical
offs between mechanical properties, material usage, and performance, and sustainability of TPU 95A components
energy efficiency. These AI-enhanced systems have been by integrating twin-screw extrusion-based filament
reported to reduce failed prints by 25%, lower energy costs production, AM using a Bambu Lab A1 3D printer, and
by 30%, and improve production efficiency by up to 15%. stress-controlled mechanical testing. Energy usage is
48
Moreover, AI is increasingly powering circular economy continuously monitored using power meters across the
practices in AM, particularly in the recycling and reuse entire workflow – from raw material preparation to final
of materials. ML algorithms are used to sort and process product – to evaluate the energy footprint associated with
waste filaments, ensuring that recycled materials maintain extrusion, printing, and testing.
optimal printability and mechanical integrity. Predictive The first stage of the study involved the preparation of
models are also used to assess the degradation of recycled TPU 95A filaments using a twin-screw extruder, a critical
polymers, allowing adaptive print settings to compensate process that determines the quality and homogeneity of
for material inconsistencies. As a result, companies have the feedstock for FDM printing. TPU 95A pellets (Dream
reported 30 – 50% reductions in the consumption of Shapes Printing, India) were loaded into the extruder
new materials through AI-enhanced recycling systems. hopper and melted at 200°C within the extrusion chamber.
In addition to energy and material efficiency, AI has The screw speed was regulated between 30 – 50 rpm to
significantly contributed to process parameter automation, ensure uniform mixing and homogenized extrusion of
enabling closed-loop control systems that dynamically the thermoplastic material. The molten TPU was extruded
adjust variables such as extrusion pressure, cooling rates, through a 1.75 mm diameter die to form a continuous
and print speed. These feedback-based systems enhance filament, which was immediately quenched in a water
print consistency while minimizing waste. AI-based bath to achieve dimensional stability, prevent defects, and
predictive maintenance also helps anticipate machine enhance material properties. The solidified filament was
failures, thus reducing unplanned downtime and resource then wound using a puller system with consistent tension.
loss. Despite these advancements, several challenges Real-time energy consumption during filament extrusion
49
remain. Many current AI models are tailored to specific was measured using a power meter, enabling detailed
machines or materials, limiting their scalability across analysis of the process efficiency and contributing to an
diverse AM platforms. Furthermore, the computational assessment of TPU 95A’s viability as a sustainable FDM
intensity of real-time monitoring and optimization material. After filament preparation, the extruded TPU
remains a barrier to widespread industrial deployment. 95A filament was processed using the Bambu Lab A1
Future efforts should focus on developing hybrid AI FDM 3D printer (China) to generate test specimens for
approaches that integrate data-driven learning with mechanical testing. The processing parameters for FDM
physics-based simulations to achieve broader applicability were optimized to balance mechanical performance, print
and more accurate, scalable optimizations. In summary, accuracy, and energy efficiency. The print speed of 70 mm/s
AI technologies have revolutionized sustainable 3D was selected to facilitate rapid printing and support
printing by reducing failure rates, optimizing material structural integrity. Nozzle temperature was maintained at
usage, and lowering energy consumption. AI-powered 230°C for optimal flow and interlayer adhesion to prevent
predictive modeling, in-process defect detection, and warping or delamination defects, while the bed temperature
adaptive control mechanisms are significantly improving was maintained at 50°C to provide optimal first-layer
the environmental sustainability of AM processes. As AI adhesion and reduce shrinkage. A layer height of 0.2 mm
continues to evolve, its role in promoting environmentally was used to enhance surface finish without compromising
and economically sustainable manufacturing within geometric accuracy, and the infill density was set at 50%
Industry 4.0 and circular manufacturing frameworks is set with a hexagonal infill pattern, chosen for its high strength-
to grow even more prominent. to-weight ratio and effective load distribution. Cooling fan

Volume 4 Issue 3 (2025) 6 doi: 10.36922/MSAM025200033