Optimizing FDM process parameters: predictive insights through taguchi, regression, and neural networks

Surface roughness (SR) is one of the major parameters used to govern the quality of the fused deposition modeling (FDM)-printed products, and the FDM process parameters can be easily regulated in order to obtain a good surface finish. The surface quality of the product produced by the FDM is generally affected by the staircase effect that needs to be managed. Also, the production time (PT) to fabricate the product and volume percentage error (VPE) should be minimized to make the FDM process more efficient. The aim of this paper is to accomplish these three objectives with the use of the parametric optimization technique integrating the artificial neural network (ANN) and the whale optimization algorithm (WOA). The FDM parameters which have been taken into consideration are layer thickness, nozzle temperature, printing speed, and raster width. Experimentation has been conducted on printed samples to examine the impact of the input parameters on SR, VPE, and PT according to Taguchi's L27 orthogonal array. The ANN model has been built up using the experimental data, which was further used as an objective function in the WOA with an aim to minimize output responses. The robustness of the proposed method has been validated on the optimal combinations of FDM process parameters.

PurposeThe purpose of the experimental investigation was to optimize the process parameters of the fused deposition modeling (FDM) technique. The optimization of the process was performed to identify the relationship between the chosen factors and the tensile strength of acrylonitrile butadiene styrene (ABS) and carbon fiber polylactic acid (PLA) thermoplastic material, FDM printed specimens. The relationship was demonstrated by using the linear experimental model analysis, and a prediction expression was established. The developed prediction expression can be used for the prediction of tensile strength of selected thermoplastic materials at a 95% confidence level.Design/methodology/approachThe Taguchi L9 experimental methodology was used to plan the total number of experiments to be performed. The process parameters were chosen as three at three working levels. The working range of chosen factors was the printing speed (60, 80 and 100mm/min), 40%, 60% and 80% as the infill density and 0.1mm, 0.2mm and 0.3mm as the layer thickness. The fused deposition modeling process parameters were optimized to get the maximum tensile strength in FDM printed ABS and carbon fiber PLA thermoplastic material specimens.FindingsThe optimum condition was achieved by the process optimization, and the desired results were obtained. The maximum desirability was achieved as 0.98 (98%) for the factors, printing speed 100mm/min, infill density 60mm and layer thickness 0.3mm. The strength of the ABS specimen was predicted to be 23.83MPa. The observed strength value was 23.66MPa. The maximum desirability was obtained as 1 (100%) for the factors, printing speed 100mm/min, infill density 60mm and layer thickness 0.2mm. The strength of the carbon fiber PLA specimen was predicted to be 26.23MPa, and the obtained value was 26.49MPa.Research limitations/implicationsThe research shows the useful process parameters and their suitable working conditions to print the tensile specimens of the ABS and carbon fiber PLA thermoplastics by using the fused deposition modeling technique. The process was optimized to identify the most influential factor, and the desired optimum condition was achieved at which the maximum tensile strength was reported. The produced prediction expression can be used to predict the tensile strength of ABS and carbon fiber PLA filaments.Practical implicationsThe results obtained from the experimental investigation are useful to get an insight into the FDM process and working limits to print the parts by using the ABS and carbon fiber PLA material for various industrial and structural applications.Social implicationsThe results will be useful in choosing the suitable thermoplastic filament for the various prototyping and structural applications. The products that require freedom in design and are difficult to produce by most of the conventional techniques can be produced at low cost and in less time by the fused deposition modeling technique.Originality/valueThe process optimization shows the practical exposures to state an optimum working condition to print the ABS and carbon fiber PLA tensile specimens by using the FDM technique. The carbon fiber PLA shows better strength than ABS thermoplastic material.

PurposeThe purpose of this paper is to investigate the mechanical properties and porosity relationships in fused deposition modelling (FDM) fabricated porous structures.Design/methodology/approachPorous structures of numerous build architectures aimed at tissue engineering (TE) application were fabricated using the FDM. The employment of FDM to fabricate these non‐random constructs offers many advantages over conventional scaffold fabrication techniques as patient specific scaffolds with well‐defined architectures and controllable pore sizes can be fabricated accurately and rapidly. There exist several FDM parameters that one needs to specify during the scaffold fabrication process. These parameters, which can be interdependent and exhibit varying effects on scaffold properties, were identified and examined using the design of experiment (DOE) approach. Essentially, the effects of five FDM process parameters, namely air gap, raster width, build orientation, build layer and build profile, on the porosity and mechanical properties of acrylonitrile‐butadienene‐styrene (ABS) scaffold structures with three‐dimensional interconnectivity were investigated in two designed experiments. Statistical analyses of the data were performed and the respective factors that have significant influence on the porosity and mechanical properties of the scaffolds were identified. The relationship between scaffold's mechanical properties and porosity was thereafter established empirically.FindingsModels of TE scaffolds of numerous build architectures were successfully fabricated using different parameter settings on the FDM. The DOE approach determined air gap and raster width as the most significant parameters in affecting the porosity and mechanical properties of the ABS scaffold structures. The relationship between scaffolds' mechanical properties and porosity was determined to be logarithmic, with the best mechanical properties observed in scaffolds of low porosity.Originality/valueThe paper highlights how the application FDM to tissue scaffold application can overcome most of the limitations encountered in the conventional techniques.

Fused Filament Fabrication (FFF), a process parameters-dependent manufacturing method, currently dominates the additive manufacturing (AM) sector because of its prominent ability to produce parts with intricate profiles, customise products, and minimise waste. Though the effects of FFF process parameters were investigated experimentally, recent research highlighted the importance of developing numerical modelling and computational methods on optimising the FFF printing process and FFF-printed materials. This study aims to investigate the tensile strength (TS) of FFF-printed high-impact polystyrene (HIPS) via devising a systematic testing and analysis framework, which combines experimental testing, representative volume element (RVE)-finite element method (FEM), rule of mixture (ROM), and artificial neural networks (ANN). HIPS samples are fabricated using FFF considering the variations of infill density, layer thickness, nozzle temperature, raster angle, and build orientation, and tested with standard tensile testing. The rule of mixtures (ROM) and its modified version (MROM) are employed to calculate the TS of longitudinally and transversely built samples at various infill densities, respectively, while an ANN model is constructed to investigate the effect of material anisotropy precisely. The optimal ANN architecture is built with five hidden layers with the number of neurons in each layer as 44, 82, 169, 362, and 50. Although both MROM and ANN perform well on the validation set, ANN exhibits superior accuracy with only a maximum error of 0.13% for training set and 11% for validation set. The combination of the RVE-FEM, MROM, and ANN approaches can significantly improve the FFF printing process of polymers for optimisation.

Fused deposition modeling (FDM) is an additive manufacturing technique used to fabricate intricate parts in 3D, within the shortest possible time without using tools, dies, fixtures, or human intervention. This article empirically reports the effects of the process parameters, i.e., the layer thickness, raster angle, raster width, air gap, part orientation, and their interactions on the accuracy of the length, width, and thickness, of acrylonitrile-butadiene-styrene (ABSP 400) parts fabricated using the FDM technique. It was found that contraction prevailed along the direction of the length and width, whereas the thickness increased from the desired value of the fabricated part. Optimum parameter settings to minimize the responses, such as the change in length, width, and thickness of the test specimen, have been determined using Taguchi’s parameter design. Because Taguchi’s philosophy fails to obtain uniform optimal factor settings for each response, in this study, a fuzzy inference system combined with the Taguchi philosophy has been adopted to generate a single response from three responses, to reach the specific target values with the overall optimum factor level settings. Further, Taguchi and artificial neural network predictive models are also presented in this study for an accuracy evaluation within the dimensions of the FDM fabricated parts, subjected to various operating conditions. The predicted values obtained from both models are in good agreement with the values from the experiment data, with mean absolute percentage errors of 3.16 and 0.15, respectively. Finally, the confirmatory test results showed an improvement in the multi-response performance index of 0.454 when using the optimal FDM parameters over the initial values.

This paper discusses the process parameters for fused deposition modelling (FDM). Layer thickness, Orientation angle and shell thickness are the process variables considered for studies. Ultimate tensile strength, dimensional accuracy and manufacturing time are the response parameters. For number of experimental runs the taguchi's L9 orthogonal array is used. Taguchis S/N ratio was used to identify a set of process parameters which give good results for respective response characteristics. Effectiveness of each parameter is investigated by using analysis of variance. The material used for the studies of process parameter is Nylon.

The global shift toward sustainable manufacturing has intensified interest in eco-friendly materials and optimized processing strategies for additive manufacturing technologies such as Fused Deposition Modeling (FDM). Polylactic acid (PLA)-based green composites have emerged as promising candidates due to their biodegradability, low environmental impact, and compatibility with FDM systems. However, the optimization of FDM process parameters for such composites remains a significant challenge due to the inherent trade-offs between mechanical performance, energy consumption, and material sustainability. This study addresses this gap by employing an integrated multi-criteria decision-making (MCDM) framework—Fuzzy Analytic Hierarchy Process (Fuzzy AHP) combined with Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)—to identify optimal FDM parameter settings for PLA-based green composites. Key process parameters, including layer thickness, print speed, infill density, and nozzle temperature, are evaluated against performance criteria such as tensile strength, surface finish, material utilization, and energy efficiency. Literature reports suggest optimal ranges such as 0.1–0.2 mm for layer thickness, 40–60 mm/s for print speed, and 80–100% for infill density to enhance part strength and minimize waste. The Fuzzy AHP–TOPSIS approach enables robust decision-making under uncertainty, providing a sustainable design methodology aligned with SDGs 4 (Quality Education), 7 (Affordable and Clean Energy), 9 (Industry, Innovation and Infrastructure), and 12 (Responsible Consumption and Production). This study establishes a foundational framework for future experimental validation and promotes informed parameter selection for sustainable, high-performance FDM manufacturing of PLA-based green composites.

Fused deposition modeling (FDM) is a widely used additive manufacturing (AM) technique for developing complex features and geometries within the shortest possible time as per customer needs. Nowadays, the customization of biomedical parts is becoming possible due to the increasing accuracy and enhanced ability of FDM machines to control the process parameters. The control on surface quality of parts produced by FDM process is of prime concern for the researchers, which are induced due to the stair steps on sloping surfaces, and needs addressing. In addition, to meet customer demands rapidly, the FDM part build time must be reduced without much compromising the strength of the parts necessary for specific applications. In the same context, this paper presents the effect and control of FDM process parameters, i.e., layer thickness, raster angle, infill density and internal structure, on surface roughness, build time and compressive strength of developed biomedical implant parts. The face-centered central composite design is employed to consummate the experimental trials, and experimental data are used to establish an adaptive neuro-fuzzy inference system (ANFIS) model for predicting the surface roughness, build time and compressive strength with respect to changes in FDM process parameters. At the end, whale optimization algorithm (WOA) has been applied to minimize surface roughness, minimize build time and maximize compressive strength simultaneously. Then, the optimal solutions obtained from WOA methodology have been compared with ANFIS predicted results. The results reveal that ANFIS-WOA methodology provides optimal combination of FDM process parameters accurately.

Additive manufacturing is gaining popularity nowadays due to its applications in the automotive, medical, aerospace industries, and sports to manufacture complex parts. Fused deposition modelling is an additive manufacturing process utilised widely due to its lower cost, fast prototyping and faster production time. The present study focuses on applying FDM to the composite material filament and the impact of parameters of FDM on the part quality. A composite filament material was prepared to combine PLA (Poly-lactic Acid) and banana fibre. Specimens were prepared using this composite filament by varying various parameters such as layer thickness, infiltration and build orientation. Flexural and tensile tests were performed as per standards. It was found that the material properties considered are significantly affected by percentage infill and build orientation. The ‘on edge’ build orientation provides better material properties as compared to the other two orientations. The tensile strength is observed to be 73% and 77% more in flat orientation and ‘on edge’ orientation, respectively, as compared to upright orientation. Also, the flat orientation and ‘on edge’ orientation respectively showed 60% and 70% more flexural strength than the upright orientation.

Fused Filament Fabrication (FFF), also known as Fused Deposition Modelling (FDM), is one of the innovative 3D printing technologies for fabricating complex components and products. Mechanical properties of 3D-printed components mostly depend on intricate process parameters of 3D printing. This study experimentally investigates the effects of four key process parameters, including layer thickness, raster angle, feed rate, and nozzle temperature, on the tensile properties and interfacial bonding behaviours of FFF printed Polylactic Acid (PLA), and their failure mechanisms. The effect of the key parameters on surface roughness is also evaluated, which is critical for enhancing manufacturing and material performance, expecting to provide a potential guide for optimisation of the FFF printing process for improving product quality. The experimental results demonstrate that tensile strength improves up to 10 and 7% with increasing nozzle temperature (200 °C to 220 °C) and low feed rate (60 mm/sec to 40 mm/sec) during the 3D printing process. The tensile strength increases up to 12% with decreasing layer thickness (0.4 mm to 0.2 mm) and 40% with decreasing raster angle (90° to 0°). The experimental findings on surface roughness indicate that FFF-printed PLA samples were significantly influenced by the layer thickness and raster angle, and an improvement in surface roughness is observed with the increase of nozzle temperature and reduction in feed rate. Microstructural SEM analysis was conducted to investigate the ruptured surfaces of the FFF printed PLA samples, focusing on the interlayer bonding quality and morphological characteristics including the effect of void formation, poor adhesion, and insufficient fusion between adjacent surface contact area with the effect of printing parameters. The feed rate and nozzle temperature were found to substantially influence the interlayer bonding between two adjacent surfaces.

Optimization of process parameters and evaluation of surface roughness for 3D printed nylon-aramid composite

Polylactic acid (PLA) is a biodegradable thermoplastic widely used in biomedical devices; however, its inherent brittleness and poor impact resistance restrict its use in mechanically demanding environments. This study investigates the mechanical, thermal, microstructural, and in vitro cytocompatibility performance of fused deposition modeling (FDM)‐printed PLA/316L stainless steel (SS) biocomposites (3–15 wt%). Composite filaments were produced by melt compounding and printed using a Taguchi‐based experimental design to assess the combined effects of reinforcement content and key FDM parameters, including raster orientation, nozzle temperature, and printing speed. Mechanical performance was evaluated through tensile, compressive, and Charpy impact testing, supported by microstructural and thermal characterization. A multi‐response Taguchi–Gray Relational Analysis (GRA) was employed to identify optimal processing conditions. Under optimal conditions (15 wt% SS, 0° raster), tensile and compressive strengths reached 94 MPa and 102 MPa, respectively, alongside a significant improvement in impact energy absorption. Thermal analyses confirmed stability within the FDM processing window, while microstructural observations revealed a transition from homogeneous particle dispersion to agglomeration‐driven defects at higher SS contents. Indirect cytotoxicity assessment following ISO 10993‐5 guidelines using L929 fibroblasts demonstrated no acute cytotoxic response for any composite formulation after 72 h of extract exposure. Overall, the findings demonstrate that SS reinforcement, combined with controlled FDM parameters, yields PLA composites with substantially improved mechanical properties while maintaining short‐term cytocompatibility, thereby supporting patient‐specific, non‐load‐bearing biomedical applications.

Experimental and statistical study on the effects of fused filament fabrication parameters on the tensile strength of hybrid PLA/Wood fabricated parts

Fused Deposition Modeling (FDM) is widely used for its cost-efficiency, design flexibility, and compatibility with advanced materials such as carbon fiber-reinforced polylactic acid (PLA-CF). This study investigates the effects of key FDM parameters layer thickness, nozzle temperature, printing speed, and raster orientation on the flexural strength, density, mass, and printing time of short PLA-carbon fiber PLA-CF components. The experimental design followed the Taguchi method, and mechanical testing was conducted using ASTM D790 three-point bending. Results show that raster orientation and layer thickness had the most significant effects on flexural strength. The optimal parameter combination was 0° raster, 0.1 mm layer, 210 °C nozzle, and 30 mm/s speed yielded the highest mechanical performance 81 MPa, along with increased density and part mass, suggesting stronger internal bonding. However, this configuration also required longer printing time. In contrast, higher layer thickness and faster speeds reduced both time and mass but led to lower strength and density. Printing speed had minimal effect on strength but impacted production efficiency. Statistical analysis using ANOVA and signal-to-noise ratios identified the most influential parameters, and residual plots confirmed model validity, showing normal distribution and constant variance. These findings provide practical insights into optimizing PLA-CF prints for structural applications by balancing mechanical strength with material efficiency and print time.

ABSTRACTThis study presents a novel approach to optimize and predict the mechanical properties of 3D‐printed polylactic acid (PLA)/wood composites through artificial neural network (ANN) and fuzzy logic (FL) modeling. The research addresses the critical challenge of determining optimal process parameters in fused deposition modeling (FDM) of natural fiber composites. Using Taguchi's L27 orthogonal array, experiments were conducted with five key printing parameters: layer thickness (100–200–300 μm), printing speed (PS) (40–60–90 mm/s), raster angle (RA) (0°–45°–90°), infill density (ID) (30%–60%–90%), and nozzle temperature (NT) (190°C–200°C–210°C). Analysis revealed that RA and PS were the most influential parameters, contributing 41.86% and 40.92% to tensile and compressive strengths, respectively. The developed ANN model demonstrated exceptional prediction accuracy with R2 values of 99.94% for both tensile and compressive strengths, surpassing the FL model's performance (R2 = 97.16%). The development of these models is crucial for accurately predicting mechanical behavior, allowing for efficient process optimization without extensive physical testing. Both methods demonstrated high prediction accuracy. Validation tests revealed that maximum errors of 1.95% and 2.81% for ANN and FL, respectively. The findings contribute valuable insights for the development of high‐performance natural fiber composites and establish a foundation for future advanced manufacturing processes.