The Effect of Variable Extrusion Parameters on the Tensile Strength and Production Quality of ABS Filaments

The Effect of Variable Extrusion Parameters on the Tensile Strength and Production Quality of ABS Filaments | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Effect of Variable Extrusion Parameters on the Tensile Strength and Production Quality of ABS Filaments Selim Bacak This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8605121/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Fused Filament Fabrication benefits from Akrilonitril Butadien Stiren due to its mechanical and thermal resistance, yet optimizing extrusion parameters for improved tensile strength and dimensional accuracy remains challenging. This study aims to analyze the impact of extrusion parameters on the tensile resistance and dimensional consistency of Akrilonitril Butadien Stiren filaments using machine learning methodologies to enhance 3D printing efficiency. Akrilonitril Butadien Stiren granules were extruded via a single-screw extruder at 225°C–245°C and screw speeds of 2–6 rpm, with the Taguchi method optimizing the experimental design. Machine learning models, including ensemble methods, were developed to predict tensile strength and filament diameter, with the Extra Trees algorithm achieving the highest accuracy (~99%). The optimal tensile strength (28.5 MPa) and filament diameter (±1.75 mm) were obtained at 235°C and 4 rpm, with temperature (81.6%) having a greater effect on tensile strength than screw speed (18.4%). This study demonstrates the successful integration of classical optimization and machine learning techniques, providing a framework for enhancing Akrilonitril Butadien Stiren filament production quality in 3D printing. 3D Printing ABS Tensile Strength Machine Learning Ensemble Methods Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 1. Introduction Fused Filament Fabrication (FFF) has been one of the cornerstones of AM technology, revolutionizing the process of prototyping and small-scale production. In contrast to traditional subtractive manufacturing, FFF builds objects layer by layer, which allows complex geometries to be made with a minimum of material waste [1, 2]. Among the various thermoplastics available for FFF fabrication, ABS has been hailed as one of the perfect engineering materials owing to its excellent impact resistance, mechanical properties, thermal stability, and relatively low cost [3, 4]. ABS has gained wide applications in the automotive industry, consumer electronics, and industrial prototypes, among others that rely on high mechanical strength and thermal stability [5,6]. The resistance of ABS to chemicals and durability also popularized it as one of the reliable engineering-grade materials [7]. Recent technology development of FFF pointed out that the optimization of extrusion parameters is crucial for quality filaments that guarantee improved printing performance [8]. Extrusion temperature, screw speed, and cooling conditions are all processing parameters that directly affect filament diameter, mechanical properties, and layer adhesion, hence affecting the overall performance of the final product [9, 10]. Some studies illustrated that optimization in those parameters can enhance the structural integrity and reliability of the 3D-printed parts considerably [11, 12]. Akrilonitril Butadien Stiren(ABS) is also considered a versatile material for functional prototype development due to its machinability and adaptability to different FFF configurations. The literature includes comprehensive analyses on how printing speed, layer height, and infill patterns influence the mechanical properties of the final product [13, 14, 15]. Different types of simulations are performed to address the challenges of production methods and to ensure industry-relevant applications. Numerical simulation usually uses the finite element method, finite difference method, and finite volume method for simulating heat transfer and fluid flow problems [16, 17, 18]. These simulations, based on applied boundary conditions, are bound to provide certain insights that can be further improved with AI and ML for more robust solutions [19]. Extrusion temperature and screw speed will be the other variables considered to study their respective effects on the quality and tensile strength of ABS filaments, thus providing valued addition to existing literature. The research paper will discuss in detail a framework of study targeted at industrial levels of improvement. 2. Materials and Methods 2.1 Materials In this work, ABS granules (KOLUMAN, White) were used as the base material. ABS is a terpolymer made during the polymerization of acrylonitrile and styrene monomers in the presence of polybutadiene, while it provides a balance of properties between toughness, heat resistance, and rigidity [20]. The granules had been dried at 90°C for 4 hours prior to processing due to manufacturers recommendations to have a homogenous extrusion with as few voids as possible in the filaments. 2.2 Methods Extrusion was done on a single-screw desktop extruder. Extrusion temperatures varied from 225°C to 245°C in steps of 5°C, while the screw speed was fixed at 2, 3, 4, 5, and 6 rpm. Filaments were fabricated at controlled ambient conditions, that is, at a room temperature of 25°C. The fabricated filaments were subsequently processed in a Snapmaker Artisan 3D printer (figure1). In order to optimize the experimental design with a minimum number of trials, the Taguchi method was employed. The parameters used in the experiment are shown in Table 1. These values were selected to analyze the effects of extrusion temperature and screw speed on filament quality. Table 1: Parameters Used in Filament Production Parameters 1 2 3 4 5 Extrusion Temperature (°C) 225 230 235 240 245 Screw Speed (rpm) 2 3 4 5 6 The Taguchi method was applied using an L25 (5²) orthogonal array with two factors (extrusion temperature and screw speed) and five levels for each parameter. This experimental design ensured that all parameter combinations were tested across a total of 25 experiments. For each trial, filaments were produced using the specified parameter combination, their diameters were measured, and tensile tests were conducted. The Signal-to-Noise (S/N) ratio was calculated for each parameter set to evaluate the robustness of the process. The primary goal was to maximize tensile strength while maintaining filament diameter consistency. The S/N ratio was calculated using the following formula: During the winding of filaments onto spools, dimensional accuracy of the filament diameter was ensured by measuring at intervals of 1 meter. These measurements were performed using a laser sensor. The optimal filament diameter was defined within a tolerance range of ±1.75 mm. Tensile test specimens were designed in a dog-bone shape according to the ASTM D638-IV [21] standard using computer-aided design (CAD) software. The specimens were fabricated using the following standard printing parameters:Extruder Temperature 250°C, Bed Temperature: 100°C, Printing Speed: 40 mm/s, Nozzle Diameter: 0.4 mm, Layer Height: 0.2 mm, Infill Density : 100%, Infill Pattern: Linear Tensile tests were conducted using a MARES universal testing machine with a 20 kN capacity at a constant speed of 1 mm/min. Tensile strength was calculated based on stress-strain relationships, providing insights into the material's performance under load. 3. Results and Discussion 3.1 Effects of Extrusion Temperature and Screw Speed on Filament Diameter The obtained filament diameters at various extrusion temperatures and screw speeds are listed in Table 2. It was observed that higher extrusion temperatures caused lower filament diameters, due to the reduced material viscosity. Conversely, higher screw speed was resulting in thicker filaments due to a larger material flow. These observations agreed well with those reported by Zou et al. [8], where the mechanical properties of ABS filaments prepared using varying conditions during extrusion were studied. The optimal diameter stability of ±1.75 mm was recorded at 235°C with 4 rpm. It means there is a very critical balance between temperature and screw speed to produce a good quality filament with consistent dimensional accuracy. Table 2. Filament Diameter at Different Parameters Screw Speed (rpm) Extrusion Temperature (°C) 225°C 230°C 235°C 240°C 245°C 2 1.82 mm 1.78 mm 1.75 mm 1.71 mm 1.68 mm 3 1.85 mm 1.80 mm 1.76 mm 1.73 mm 1.70 mm 4 1.88 mm 1.83 mm 1.77 mm 1.75 mm 1.72 mm 5 1.91 mm 1.87 mm 1.79 mm 1.76 mm 1.74 mm 6 1.93 mm 1.90 mm 1.81 mm 1.78 mm 1.75 mm Figure 2 illustrates the variations in tensile strength with respect to extrusion temperatures at different screw speeds. The figure also highlights the importance of balanced parameters in enhancing mechanical properties. 3.2 Effects of Extrusion Temperature and Screw Speed on Tensile Strength Table 3: Values of ultimate tensile strength, MPa, obtained at different screw speed-extrusion temperature combinations. The minimum value was recorded at a tensile strength of 23.5 MPa at 245°C and 2 rpm screw speed, while the highest recorded tensile strength was 28.5 MPa at 235°C and 4 rpm. Too low a temperature provides poor layer adhesion; too high a temperature initiates degradation and thus affects mechanical performance negatively. These findings confirm the work done by Subramaniam et al. [9], where optimal extrusion temperatures are realized to get superior mechanical properties. Precise control of the extrusion parameters results in a significant improvement in the mechanical integrity of ABS parts. Table 3. Ultimate Tensile Strength at Different Parameters Screw Speed (rpm) Extrusion Temperature (°C) 225°C 230°C 235°C 240°C 245°C 2 24.1 MPa 25.3 MPa 26.4 MPa 24.9 MPa 23.5 MPa 3 25.0 MPa 26.1 MPa 27.2 MPa 25.7 MPa 24.3 MPa 4 26.7 MPa 27.4 MPa 28.5 MPa 27.1 MPa 25.8 MPa 5 27.1 MPa 27.8 MPa 28.3 MPa 27.5 MPa 26.0 MPa 6 26.3 MPa 26.9 MPa 27.6 MPa 26.8 MPa 25.4 MPa Nevertheless, the tensile strength at a temperature of 225°C to 230°C was somewhat lower than that within the range normally measured for commercial ABS filaments, ranging between 24.1 and 26.9 MPa. The results obtained at 245°C demonstrated the mechanical weakening due to the thermal degradation effect. More striking is that one of the smallest tensile strengths, 23.5 MPa, has been obtained with the combination of 245°C and 2 rpm screw speed. These findings have pointed out that both the extrusion temperature and screw speed significantly affect the mechanical properties of the filament. While higher temperatures (e.g., 235°C and 240°C) optimize the material's mechanical performance, excessively high temperatures (e.g., 245°C) may lead to thermal degradation, resulting in cracks and brittleness. This trend is illustrated in Figure 3. Figure 4 visually represents the effects of temperature and speed combinations on tensile strength, thereby supporting these findings. The heatmap illustrates the relationship between tensile strength, extrusion speed, and temperature, highlighting the optimal combinations for maximum performance. It is observed that the tensile strength reaches its highest values (approximately 28 MPa) in the region where the extrusion temperature is around 235 °C and the screw speed is approximately 4.5 rpm. Outside this region, an increase or decrease in temperature and speed results in a drop in tensile strength. Notably, the tensile strength reaches its lowest levels (23.5 MPa) at low temperature and low screw speed. These results demonstrate the strong influence of parameters on tensile strength and indicate that optimum performance is achieved within a specific range of temperature and speed. Such analyses are critical for optimizing material and production processes. [17] 3.3 Machine Learning and Ensemble Methods Analysis Results The investigated effects of temperature and screw speed on tensile strength with various machine learning algorithms will now be presented. In this respect, while temperature and screw speed are independent variables, the tensile strength was taken as the dependent variable, and accordingly, data were modeled. The analyses were done using various machine learning techniques to have a deeper understanding of the effect of the parameters on tensile strength. Algorithms such as AdaBoost, Extra Trees, Support Vector Regression, Gradient Boosting, and Random Forest were employed to evaluate each model's capacity to fit the dataset, error rates, and accuracy levels, summarized in Table 4. Table 4. Analysis Results of the Models Used Algorithm MSE RMSE R-Squared Accuracy (%) AdaBoost 0.105007903 0.32404923 0.965762789 98.97970642 Extra Trees 1.23E-06 0.0011094 0.999999703 99.99941503 SVR 0.0080483 0.089712315 0.9980575 99.6568392 Gradient Boosting 0.006142301 0.078372836 0.998579212 99.75652679 Random Forest 0.017186981 0.131099127 0.996024445 99.54453239 The AdaBoost analysis (Figure 5) demonstrated a very high performance with an accuracy rate of 98.98%. However, in terms of errors, the MSE (0.1050) and RMSE (0.3240) values are relatively high. The R-squared value (0.9658) indicates that the model has a strong explanatory power but provides less effective predictions compared to other algorithms. The Extra Trees analysis achieved the best results (Figure 6), demonstrating excellent performance with MSE (1.23e-06) and RMSE (0.0011) values. The accuracy rate of this model (99.999%) indicates an almost error-free prediction. Furthermore, the R-squared value (1.00) confirms the model's perfect explanatory power. Compared to other algorithms, Extra Trees shows a clear superiority in both accuracy and error rates. The analysis using the SVR algorithm (Figure 6) demonstrated notable performance with MSE (0.0080) and RMSE (0.0897) values. The R-squared value (0.9981) indicates a high explanatory power of the model, while the accuracy rate of 99.66% reflects a strong performance. The Gradient Boosting model (Figure 7) achieved low error rates with MSE (0.0061) and RMSE (0.0783) values. The R-squared value (0.9986) and accuracy rate (99.76%) indicate that the model has strong overall performance. Compared to AdaBoost, Gradient Boosting provides lower error rates and demonstrates superiority in terms of accuracy. Lastly, the Random Forest analysis results (Figure 9) demonstrated good performance with MSE (0.0172) and RMSE (0.1311) values. The R-squared value (0.9960) and accuracy rate (99.54%) indicate that the model possesses high predictive accuracy. However, compared to Gradient Boosting and Extra Trees, the error rates of Random Forest are slightly higher. Additionally, the feature importance ranking from this model revealed the effects of extrusion parameters, temperature, and speed on tensile strength. It was concluded that temperature (%81.6) is a more influential parameter than speed (%18.4). Table 5 presents the hyperparameter settings of five different algorithms in detail. In the AdaBoost algorithm, the number of estimators ('n_estimators') and learning rate ('learning_rate') parameters were optimized over various value ranges. The Extra Trees algorithm required the optimization of multiple hyperparameters, such as the number of decision trees ('n_estimators'), maximum depth ('max_depth'), minimum samples for splitting ('min_samples_split'), and minimum samples per leaf ('min_samples_leaf'). For the SVR algorithm, the kernel type ('kernel'), regularization parameter ('C'), and gamma value were fixed. Gradient Boosting stood out with adjustments to various learning rates ('learning_rate'), numbers of estimators, tree depths, and minimum split and leaf sizes. Lastly, Random Forest was tuned using parameters such as the number of estimators, maximum depth, and the minimum number of splits and leaves. These different configurations were implemented to maximize the performance of each algorithm based on the characteristics of the dataset. Table 5. Hyperparameter Settings of Algorithms Algorithms Hyperparameters Values AdaBoost 'n_estimators': [30, 50, 75, 100, 200], 'learning_rate': [0.05, 0.1, 0.5, 1, 2] Extra Trees 'n_estimators': [50, 100, 200], 'max_depth': [None, 10, 20, 30], 'min_samples_split': [2, 5, 10], 'min_samples_leaf': [1, 2, 4] SVR 'kernel': 'rbf', 'C': 1e3, 'gamma': 0.1 Gradient Boosting 'n_estimators': [50, 75, 100], 'max_depth': [2, 4, 8], 'min_samples_split': [1, 2, 4], 'min_samples_leaf': [1, 2, 8], 'learning_rate': [0.05, 0.1, 0.5, 1] Random Forest 'n_estimators': [50, 100, 200], 'max_depth': [2, 4, 8, 16], 'min_samples_split': [2, 3, 4], 'min_samples_leaf': [2, 3, 4, 8] 4. Conclusion Various parameters in filament extrusion directly influence the produced filaments and, eventually, the 3D-printed samples. The optimized and accurate combination of these parameters allows producing high-quality filaments that improve several aspects of the performance of 3D parts. In this study, the five different extrusion temperatures, ranging from 225°C to 245°C, and five different extrusion speeds from 2 rpm to 6 rpm, were optimized using the Taguchi method. This range of temperature, 230°C to 240°C, and speed, 4 to 5 rpm, yielded the most consistent and strongest results in terms of tensile strength. In this combination, the highest value for tensile strength reached around 26 MPa to 28 MPa. Filament diameter received within this parameter combination had consistency in diameter at about 1.75 mm to 1.77 mm. In this study, using the DOE technique, it was able to optimize the extrusion process parameters for the production of high-strength 3D printing filaments while keeping the filament diameter within the set limits of tolerance. The DOE technique not only optimized the extrusion parameters but also provided a robust foundation in enhancing repeatability and quality of the filament production process. Future work might address the application of DOE to printing parameters and also extend to the adaptation of these to different types of material. The effects of temperature and speed over the long term on material properties can be examined in relation to the stress-relaxation and fatigue resistance. The analyses were done using machine learning and ensemble methods with five different models. Among them, the best validation accuracy of 99.9% was from the Extra Trees algorithm. This algorithm outperformed others in both accuracy and error rates. Further research may serve to test other types of thermoplastic materials in light of the findings presented in this work, together with the possibility of taking into account the effects on specimens with more complicated geometrical shapes, which may be of greater value for the possible use in industry. Declarations Author Contributions Selim Bacak: Conceptualization; methodology; software; formal analysis; resources; writing – original draft; writing – review and editing. Conflict of Interest Statement The author declare that there is no conflict of interest in the preparation and publication of this study. Funding Statement This study did not receive any external funding support. References Sheoran, A. J., & Kumar, H. (2020). Optimization of FDM process parameters. Materials Today: Proceedings, 21, 1659–1672. Jandyal, A., et al. (2022). Review of processes, materials, and applications in FFF. Sustainable Operations and Computers, 3, 33–42. Fontana, L., et al. (2022). Influence of 3D printing parameters on tensile strength. Materials Today: Proceedings, 57, 657–663. I. Gibson, D. Rosen, B. Stucker, M. Khorasani, Development of additive manufacturing technology, in: Additive Manufacturing Technologies, Springer, Cham, 2021, https://doi.org/10.1007/978-3-030-56127-7_2.. S. Kumar, I. Singh, S.S.R. Koloor, D. Kumar, M.Y. Yahya, On laminated object manufactured FDM-printed ABS/TPU multimaterial specimens: an insight into mechanical and morphological characteristics, Polymers 14 (19) (2022) 4066, A.R.T. Perez, D.A. Roberson, R.B. Wicker, Fracture surface analysis of 3D-Printed tensile specimens of novel ABS-Based materials, J. Fail. Anal. Prev. 14 (3) (2014) 343–353, https://doi.org/10.1007/s11668-014-9803-9. Chacón, J. M., et al. (2017). Additive manufacturing of ABS: Mechanical properties and optimization. Material Design, 124, 143–157. Zou, R., et al. (2016). Elasticity and mechanical response of 3D printed materials. Composites Part B, 99, 506–513. Subramaniam, S. R., et al. (2019). Evaluating ABS tensile strength with varying parameters. AIP Conference Proceedings, 2059. Birosz, Márton Tamás and Andó, Mátyás and Ledenyák, Dániel, Effect of FDM Infill Patterns on Mechanical Properties. Available at SSRN: https://ssrn.com/abstract=3950131 or http://dx.doi.org/10.2139/ssrn.3950131 Ahmed, Syed Waqar, Ghulam Hussain, Khurram Altaf, Sadaqat Ali, Mohammed Alkahtani, Mustufa Haider Abidi, and Ayoub Alzabidi. On the effects of process parameters and optimization of interlaminate bond strength in 3D printed ABS/CF-PLA composite. Polymers, 12, 9, (2020) 2155 NatureWorks. (2018). ABS technical data sheet. Retrieved from https://www.natureworksllc.com. Chen, K., et al. (2021). Optimization of printing parameters for ABS composites. Thin-Walled Structures, 164, 107717. M. Mirzaee, S. Noghanian, L. Wiest, I. Chang, July). Developing flexible 3D printed antenna using conductive ABS materials, in: 2015 IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting, 2015 S. Ahn, M. Montero, D. Odell, S. Roundy, P.K. Wright, Anisotropic material properties of fused deposition modeling ABS, Rapid Prototyp. J. 8 (4) (2002) 248–257, https://doi.org/10.1108/13552540210441166. Tichý, T.; Šefl, O.; Veselý, P.; Dušek, K.; Bušek, D. Mathematical modelling of temperature distribution in selected parts of fff printer during 3d printing process. Polymers 2021, 13, 4213. Ramos, N.; Mittermeier, C.; Kiendl, J. Experimental and numerical investigations on heat transfer in fused filament fabrication 3D-printed specimens. Int. J. Adv. Manuf. Technol. 2022, 118, 1367–1381. Phan, D.D.; Horner, J.S.; Swain, Z.R.; Beris, A.N.; Mackay, M.E. Computational fluid dynamics simulation of the melting process in the fused filament fabrication additive manufacturing technique. Addit. Manuf. 2020, 33, 101161. Zhu, Q.; Liu, Z.; Yan, J. Machine learning for metal additive manufacturing: Predicting temperature and melt pool fluid dynamics using physics-informed neural networks. Comput. Mech. 2021, 67, 619–635. M. Galetto, E. Verna, G. Genta, Effect of process parameters on parts quality and process efficiency of fused deposition modeling, Computer and Industrial engineering 156 (2021) 107238, https://doi.org/10.1016/j.cie.2021.107238. Z. Weng, J. Wang, T. Senthil, L. Wu, Mechanical and thermal properties of ABS/ montmorillonite nanocomposites for fused deposition modeling 3D printing, Mater. Des. 102 (2016) 276–283, https://doi.org/10.1016/j. matdes.2016.04.045. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8605121","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":575920384,"identity":"582d19cc-0f48-4c1a-bdc2-45d5d7c75642","order_by":0,"name":"Selim Bacak","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABD0lEQVRIie3RsUoDMRjA8YTA3ZLj1hTFvMJ3kwhSXyVyc3FzKdgrQrqcuHZQfIW63ZgzYJfUrnW74qpgcTpczFlaFLzQ0SH/IYGQH3whCPl8/zKSKZzZnRKltofUSfCGBEJtb9uFuQhaEwpoJ8JHw+HDc9HlcW5WZV10z05Chas3iS4Os78JmDLTPZMm49nVREcmPcqpIMmtRGxftRB2aokkAs2jicaSAEUi2IssaZuM3y0bMhB8TquylgOgcRV+ugha4IZoAbMcqUhqoEwExEXANIOZaXJvHsG+ZQp0sbzs3DyxzrhtsJHWH72izw9M+rKqiz6E12n5/np+HDs+5kffH7RedwMb4vP5fL7ffQHx118/UvX+GQAAAABJRU5ErkJggg==","orcid":"","institution":"Isparta Applied Science University, Technical Vocational High School","correspondingAuthor":true,"prefix":"","firstName":"Selim","middleName":"","lastName":"Bacak","suffix":""}],"badges":[],"createdAt":"2026-01-14 21:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8605121/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8605121/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100684456,"identity":"6d654892-b14e-40da-9ead-d1473afcabd6","added_by":"auto","created_at":"2026-01-20 12:43:15","extension":"doc","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":716645,"visible":true,"origin":"","legend":"","description":"","filename":"THEEFF1.doc","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/67a5cc8a0d5eeadb3f64c707.doc"},{"id":100684448,"identity":"3528c61e-2b5f-4805-8128-67a332d58710","added_by":"auto","created_at":"2026-01-20 12:43:10","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3024,"visible":true,"origin":"","legend":"","description":"","filename":"b510918967824152bf3fc702dbddf386.json","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/86068513ca4b00223dfbaa49.json"},{"id":100684872,"identity":"d4ddfebe-dfcb-42f3-8077-ffed8a8aba04","added_by":"auto","created_at":"2026-01-20 12:47:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":247950,"visible":true,"origin":"","legend":"\u003cp\u003eSample samples\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/e4ee7d48c92b53e69492230e.png"},{"id":100684488,"identity":"4fb2bf0f-1ca6-4ffc-90cb-d6d999436cd0","added_by":"auto","created_at":"2026-01-20 12:43:29","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":278731,"visible":true,"origin":"","legend":"\u003cp\u003eVariations in Tensile Strength with Respect to Extrusion Temperatures at Different Screw Speeds\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/ae47a1655679be208f91adf2.png"},{"id":100684614,"identity":"94ba1083-7776-45d8-ba0f-35d92525b9e5","added_by":"auto","created_at":"2026-01-20 12:44:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":267699,"visible":true,"origin":"","legend":"\u003cp\u003eTensile Test Results of Samples at 235°C with Different Extrusion Speeds\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/6ad1df3e9429bf46c88f706f.png"},{"id":100684664,"identity":"4840d977-b3c8-486f-bc44-d165d85c38cc","added_by":"auto","created_at":"2026-01-20 12:45:18","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":306139,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmap of Tensile Strength Among Parameters\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/d16905f7e93875c9efe8d405.png"},{"id":100684650,"identity":"438809fa-c24d-42d0-a28b-09e16db1c0db","added_by":"auto","created_at":"2026-01-20 12:45:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":103166,"visible":true,"origin":"","legend":"\u003cp\u003eAdaBoost Analysis Results\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/635e789cda3414b0ebd6f5ce.png"},{"id":100685095,"identity":"f8774053-4ec8-44e4-ab30-d876b30d0d15","added_by":"auto","created_at":"2026-01-20 12:49:07","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":114966,"visible":true,"origin":"","legend":"\u003cp\u003eExtra Tress Analysis Results\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/f3f0d0a13b11ace792f4eff5.png"},{"id":100684565,"identity":"34531949-b856-4788-998d-1e583f0afcea","added_by":"auto","created_at":"2026-01-20 12:44:07","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":115293,"visible":true,"origin":"","legend":"\u003cp\u003eSVR Analysis Results\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/e78a1dbd4d19a63efb6b9950.png"},{"id":100684863,"identity":"16d64aac-30df-4553-a4fd-cdebe7e41257","added_by":"auto","created_at":"2026-01-20 12:47:01","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":101516,"visible":true,"origin":"","legend":"\u003cp\u003eGradient Boosting Analysis Results\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/4bc2da33f6a74c815b950ae1.png"},{"id":100684688,"identity":"b3125e62-6634-4222-94c0-9398754b7cd4","added_by":"auto","created_at":"2026-01-20 12:45:32","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":109994,"visible":true,"origin":"","legend":"\u003cp\u003eRandom Forest Analysis Results\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/eb691c36ac4c56a4615e89d2.png"},{"id":101226675,"identity":"61f27743-8373-4c58-b0af-0a6f8d5bf629","added_by":"auto","created_at":"2026-01-27 12:55:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2700391,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8605121/v1/8f4bbb25-0b53-4c9e-8dbb-8674b86b6760.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eThe Effect of Variable Extrusion Parameters on the Tensile Strength and Production Quality of ABS Filaments\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eFused Filament Fabrication (FFF) has been one of the cornerstones of AM technology, revolutionizing the process of prototyping and small-scale production. In contrast to traditional subtractive manufacturing, FFF builds objects layer by layer, which allows complex geometries to be made with a minimum of material waste [1, 2]. Among the various thermoplastics available for FFF fabrication, ABS has been hailed as one of the perfect engineering materials owing to its excellent impact resistance, mechanical properties, thermal stability, and relatively low cost [3, 4].\u003c/p\u003e\n\u003cp\u003eABS has gained wide applications in the automotive industry, consumer electronics, and industrial prototypes, among others that rely on high mechanical strength and thermal stability [5,6]. The resistance of ABS to chemicals and durability also popularized it as one of the reliable engineering-grade materials [7]. Recent technology development of FFF pointed out that the optimization of extrusion parameters is crucial for quality filaments that guarantee improved printing performance [8].\u003c/p\u003e\n\u003cp\u003eExtrusion temperature, screw speed, and cooling conditions are all processing parameters that directly affect filament diameter, mechanical properties, and layer adhesion, hence affecting the overall performance of the final product [9, 10]. Some studies illustrated that optimization in those parameters can enhance the structural integrity and reliability of the 3D-printed parts considerably [11, 12].\u003c/p\u003e\n\u003cp\u003eAkrilonitril Butadien Stiren(ABS) is also considered a versatile material for functional prototype development due to its machinability and adaptability to different FFF configurations. The literature includes comprehensive analyses on how printing speed, layer height, and infill patterns influence the mechanical properties of the final product [13, 14, 15].\u003c/p\u003e\n\u003cp\u003eDifferent types of simulations are performed to address the challenges of production methods and to ensure industry-relevant applications. Numerical simulation usually uses the finite element method, finite difference method, and finite volume method for simulating heat transfer and fluid flow problems [16, 17, 18]. These simulations, based on applied boundary conditions, are bound to provide certain insights that can be further improved with AI and ML for more robust solutions [19].\u003c/p\u003e\n\u003cp\u003eExtrusion temperature and screw speed will be the other variables considered to study their respective effects on the quality and tensile strength of ABS filaments, thus providing valued addition to existing literature. The research paper will discuss in detail a framework of study targeted at industrial levels of improvement.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003e2.1 Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this work, ABS granules (KOLUMAN, White) were used as the base material. ABS is a terpolymer made during the polymerization of acrylonitrile and styrene monomers in the presence of polybutadiene, while it provides a balance of properties between toughness, heat resistance, and rigidity [20]. The granules had been dried at 90\u0026deg;C for 4 hours prior to processing due to manufacturers recommendations to have a homogenous extrusion with as few voids as possible in the filaments.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExtrusion was done on a single-screw desktop extruder. Extrusion temperatures varied from 225\u0026deg;C to 245\u0026deg;C in steps of 5\u0026deg;C, while the screw speed was fixed at 2, 3, 4, 5, and 6 rpm. Filaments were fabricated at controlled ambient conditions, that is, at a room temperature of 25\u0026deg;C. The fabricated filaments were subsequently processed in a Snapmaker Artisan 3D printer (figure1). In order to optimize the experimental design with a minimum number of trials, the Taguchi method was employed.\u003c/p\u003e\n\u003cp\u003eThe parameters used in the experiment are shown in Table 1. These values were selected to analyze the effects of extrusion temperature and screw speed on filament quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1:\u0026nbsp;\u003c/strong\u003eParameters Used in Filament Production\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"501\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eParameters\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 74px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 180px;\"\u003e\n \u003cp\u003eExtrusion Temperature (\u0026deg;C)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e225\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 74px;\"\u003e\n \u003cp\u003e230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e245\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 180px;\"\u003e\n \u003cp\u003eScrew Speed (rpm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 74px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eThe Taguchi method was applied using an L25 (5\u0026sup2;) orthogonal array with two factors (extrusion temperature and screw speed) and five levels for each parameter. This experimental design ensured that all parameter combinations were tested across a total of 25 experiments. For each trial, filaments were produced using the specified parameter combination, their diameters were measured, and tensile tests were conducted.\u003c/p\u003e\n\u003cp\u003eThe Signal-to-Noise (S/N) ratio was calculated for each parameter set to evaluate the robustness of the process. The primary goal was to maximize tensile strength while maintaining filament diameter consistency.\u003c/p\u003e\n\u003cp\u003eThe S/N ratio was calculated using the following formula:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"250\" height=\"61\"\u003e\u003c/p\u003e\n\u003cp\u003eDuring the winding of filaments onto spools, dimensional accuracy of the filament diameter was ensured by measuring at intervals of 1 meter. These measurements were performed using a laser sensor. The optimal filament diameter was defined within a tolerance range of \u0026plusmn;1.75 mm.\u003c/p\u003e\n\u003cp\u003eTensile test specimens were designed in a dog-bone shape according to the ASTM D638-IV [21] standard using computer-aided design (CAD) software. The specimens were fabricated using the following standard printing parameters:Extruder Temperature 250\u0026deg;C, Bed Temperature: 100\u0026deg;C, Printing Speed: 40 mm/s, Nozzle Diameter: 0.4 mm, Layer Height: 0.2 mm, Infill Density\u003cstrong\u003e:\u003c/strong\u003e 100%, Infill Pattern: Linear\u003c/p\u003e\n\u003cp\u003eTensile tests were conducted using a MARES universal testing machine with a 20 kN capacity at a constant speed of 1 mm/min. Tensile strength was calculated based on stress-strain relationships, providing insights into the material\u0026apos;s performance under load.\u003c/p\u003e"},{"header":"3. Results and Discussion","content":"\u003cp\u003e\u003cstrong\u003e3.1 Effects of Extrusion Temperature and Screw Speed on Filament Diameter\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe obtained filament diameters at various extrusion temperatures and screw speeds are listed in Table 2. It was observed that higher extrusion temperatures caused lower filament diameters, due to the reduced material viscosity. Conversely, higher screw speed was resulting in thicker filaments due to a larger material flow. These observations agreed well with those reported by Zou et al. [8], where the mechanical properties of ABS filaments prepared using varying conditions during extrusion were studied.\u003c/p\u003e\n\u003cp\u003eThe optimal diameter stability of \u0026plusmn;1.75 mm was recorded at 235\u0026deg;C with 4 rpm. It means there is a very critical balance between temperature and screw speed to produce a good quality filament with consistent dimensional accuracy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2.\u0026nbsp;\u003c/strong\u003eFilament Diameter at Different Parameters\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"3\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScrew Speed (rpm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eExtrusion Temperature (\u0026deg;C)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e225\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e230\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e235\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e240\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e245\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.82 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.78 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.75 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.71 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.68 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.85 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.80 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.76 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.73 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.70 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.88 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.83 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.77 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.75 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.72 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.91 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.87 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.79 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.76 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.74 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.93 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.90 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.81 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.78 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.75 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 2 illustrates the variations in tensile strength with respect to extrusion temperatures at different screw speeds. The figure also highlights the importance of balanced parameters in enhancing mechanical properties.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Effects of Extrusion Temperature and Screw Speed on Tensile Strength\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 3: Values of ultimate tensile strength, MPa, obtained at different screw speed-extrusion temperature combinations. The minimum value was recorded at a tensile strength of 23.5 MPa at 245\u0026deg;C and 2 rpm screw speed, while the highest recorded tensile strength was 28.5 MPa at 235\u0026deg;C and 4 rpm. Too low a temperature provides poor layer adhesion; too high a temperature initiates degradation and thus affects mechanical performance negatively.\u003c/p\u003e\n\u003cp\u003eThese findings confirm the work done by Subramaniam et al. [9], where optimal extrusion temperatures are realized to get superior mechanical properties. Precise control of the extrusion parameters results in a significant improvement in the mechanical integrity of ABS parts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3.\u0026nbsp;\u003c/strong\u003eUltimate Tensile Strength at Different Parameters\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"3\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScrew Speed (rpm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eExtrusion Temperature (\u0026deg;C)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e225\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e230\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e235\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e240\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e245\u0026deg;C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24.1 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25.3 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.4 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24.9 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e23.5 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25.0 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.1 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.2 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25.7 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24.3 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.7 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.4 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e28.5 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.1 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25.8 MPa\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.1 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.8 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e28.3 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.5 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.0 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.3 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.9 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.6 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e26.8 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e25.4 MPa\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eNevertheless, the tensile strength at a temperature of 225\u0026deg;C to 230\u0026deg;C was somewhat lower than that within the range normally measured for commercial ABS filaments, ranging between 24.1 and 26.9 MPa. The results obtained at 245\u0026deg;C demonstrated the mechanical weakening due to the thermal degradation effect. More striking is that one of the smallest tensile strengths, 23.5 MPa, has been obtained with the combination of 245\u0026deg;C and 2 rpm screw speed.\u003c/p\u003e\n\u003cp\u003eThese findings have pointed out that both the extrusion temperature and screw speed significantly affect the mechanical properties of the filament. While higher temperatures (e.g., 235\u0026deg;C and 240\u0026deg;C) optimize the material\u0026apos;s mechanical performance, excessively high temperatures (e.g., 245\u0026deg;C) may lead to thermal degradation, resulting in cracks and brittleness. This trend is illustrated in Figure 3.\u003c/p\u003e\n\u003cp\u003eFigure 4 visually represents the effects of temperature and speed combinations on tensile strength, thereby supporting these findings.\u003c/p\u003e\n\u003cp\u003eThe heatmap illustrates the relationship between tensile strength, extrusion speed, and temperature, highlighting the optimal combinations for maximum performance. It is observed that the tensile strength reaches its highest values (approximately 28 MPa) in the region where the extrusion temperature is around 235 \u0026deg;C and the screw speed is approximately 4.5 rpm. Outside this region, an increase or decrease in temperature and speed results in a drop in tensile strength. Notably, the tensile strength reaches its lowest levels (23.5 MPa) at low temperature and low screw speed. These results demonstrate the strong influence of parameters on tensile strength and indicate that optimum performance is achieved within a specific range of temperature and speed. Such analyses are critical for optimizing material and production processes. [17]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Machine Learning and Ensemble Methods Analysis Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe investigated effects of temperature and screw speed on tensile strength with various machine learning algorithms will now be presented. In this respect, while temperature and screw speed are independent variables, the tensile strength was taken as the dependent variable, and accordingly, data were modeled.\u003c/p\u003e\n\u003cp\u003eThe analyses were done using various machine learning techniques to have a deeper understanding of the effect of the parameters on tensile strength. Algorithms such as AdaBoost, Extra Trees, Support Vector Regression, Gradient Boosting, and Random Forest were employed to evaluate each model\u0026apos;s capacity to fit the dataset, error rates, and accuracy levels, summarized in Table 4.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4.\u0026nbsp;\u003c/strong\u003eAnalysis Results of the Models Used\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlgorithm\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMSE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRMSE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eR-Squared\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAccuracy (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdaBoost\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.105007903\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.32404923\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.965762789\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e98.97970642\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eExtra Trees\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e1.23E-06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.0011094\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.999999703\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e99.99941503\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSVR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.0080483\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.089712315\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.9980575\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e99.6568392\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGradient Boosting\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.006142301\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.078372836\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.998579212\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e99.75652679\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 25px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandom Forest\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.017186981\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.131099127\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e0.996024445\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 18px;\"\u003e\n \u003cp\u003e99.54453239\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe AdaBoost analysis (Figure 5) demonstrated a very high performance with an accuracy rate of 98.98%. However, in terms of errors, the MSE (0.1050) and RMSE (0.3240) values are relatively high. The R-squared value (0.9658) indicates that the model has a strong explanatory power but provides less effective predictions compared to other algorithms.\u003c/p\u003e\n\u003cp\u003eThe Extra Trees analysis achieved the best results (Figure 6), demonstrating excellent performance with MSE (1.23e-06) and RMSE (0.0011) values. The accuracy rate of this model (99.999%) indicates an almost error-free prediction. Furthermore, the R-squared value (1.00) confirms the model\u0026apos;s perfect explanatory power. Compared to other algorithms, Extra Trees shows a clear superiority in both accuracy and error rates.\u003c/p\u003e\n\u003cp\u003eThe analysis using the SVR algorithm (Figure 6) demonstrated notable performance with MSE (0.0080) and RMSE (0.0897) values. The R-squared value (0.9981) indicates a high explanatory power of the model, while the accuracy rate of 99.66% reflects a strong performance.\u003c/p\u003e\n\u003cp\u003eThe Gradient Boosting model (Figure 7) achieved low error rates with MSE (0.0061) and RMSE (0.0783) values. The R-squared value (0.9986) and accuracy rate (99.76%) indicate that the model has strong overall performance. Compared to AdaBoost, Gradient Boosting provides lower error rates and demonstrates superiority in terms of accuracy.\u003c/p\u003e\n\u003cp\u003eLastly, the Random Forest analysis results (Figure 9) demonstrated good performance with MSE (0.0172) and RMSE (0.1311) values. The R-squared value (0.9960) and accuracy rate (99.54%) indicate that the model possesses high predictive accuracy. However, compared to Gradient Boosting and Extra Trees, the error rates of Random Forest are slightly higher. Additionally, the feature importance ranking from this model revealed the effects of extrusion parameters, temperature, and speed on tensile strength. It was concluded that temperature (%81.6) is a more influential parameter than speed (%18.4).\u003c/p\u003e\n\u003cp\u003eTable 5 presents the hyperparameter settings of five different algorithms in detail. In the AdaBoost algorithm, the number of estimators (\u0026apos;n_estimators\u0026apos;) and learning rate (\u0026apos;learning_rate\u0026apos;) parameters were optimized over various value ranges. The Extra Trees algorithm required the optimization of multiple hyperparameters, such as the number of decision trees (\u0026apos;n_estimators\u0026apos;), maximum depth (\u0026apos;max_depth\u0026apos;), minimum samples for splitting (\u0026apos;min_samples_split\u0026apos;), and minimum samples per leaf (\u0026apos;min_samples_leaf\u0026apos;). For the SVR algorithm, the kernel type (\u0026apos;kernel\u0026apos;), regularization parameter (\u0026apos;C\u0026apos;), and gamma value were fixed. Gradient Boosting stood out with adjustments to various learning rates (\u0026apos;learning_rate\u0026apos;), numbers of estimators, tree depths, and minimum split and leaf sizes. Lastly, Random Forest was tuned using parameters such as the number of estimators, maximum depth, and the minimum number of splits and leaves. These different configurations were implemented to maximize the performance of each algorithm based on the characteristics of the dataset.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 5.\u0026nbsp;\u003c/strong\u003eHyperparameter Settings of Algorithms\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"612\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlgorithms\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHyperparameters Values\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdaBoost\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u0026apos;n_estimators\u0026apos;: [30, 50, 75, 100, 200], \u0026apos;learning_rate\u0026apos;: [0.05, 0.1, 0.5, 1, 2]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eExtra Trees\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u0026apos;n_estimators\u0026apos;: [50, 100, 200], \u0026apos;max_depth\u0026apos;: [None, 10, 20, 30], \u0026apos;min_samples_split\u0026apos;: [2, 5, 10], \u0026apos;min_samples_leaf\u0026apos;: [1, 2, 4]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSVR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u0026apos;kernel\u0026apos;: \u0026apos;rbf\u0026apos;, \u0026apos;C\u0026apos;: 1e3, \u0026apos;gamma\u0026apos;: 0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGradient Boosting\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u0026apos;n_estimators\u0026apos;: [50, 75, 100], \u0026apos;max_depth\u0026apos;: [2, 4, 8], \u0026apos;min_samples_split\u0026apos;: [1, 2, 4], \u0026apos;min_samples_leaf\u0026apos;: [1, 2, 8], \u0026apos;learning_rate\u0026apos;: [0.05, 0.1, 0.5, 1]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandom Forest\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 485px;\"\u003e\n \u003cp\u003e\u0026apos;n_estimators\u0026apos;: [50, 100, 200], \u0026apos;max_depth\u0026apos;: [2, 4, 8, 16], \u0026apos;min_samples_split\u0026apos;: [2, 3, 4], \u0026apos;min_samples_leaf\u0026apos;: [2, 3, 4, 8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eVarious parameters in filament extrusion directly influence the produced filaments and, eventually, the 3D-printed samples. The optimized and accurate combination of these parameters allows producing high-quality filaments that improve several aspects of the performance of 3D parts. In this study, the five different extrusion temperatures, ranging from 225\u0026deg;C to 245\u0026deg;C, and five different extrusion speeds from 2 rpm to 6 rpm, were optimized using the Taguchi method. This range of temperature, 230\u0026deg;C to 240\u0026deg;C, and speed, 4 to 5 rpm, yielded the most consistent and strongest results in terms of tensile strength. In this combination, the highest value for tensile strength reached around 26 MPa to 28 MPa. Filament diameter received within this parameter combination had consistency in diameter at about 1.75 mm to 1.77 mm.\u003c/p\u003e\n\u003cp\u003eIn this study, using the DOE technique, it was able to optimize the extrusion process parameters for the production of high-strength 3D printing filaments while keeping the filament diameter within the set limits of tolerance. The DOE technique not only optimized the extrusion parameters but also provided a robust foundation in enhancing repeatability and quality of the filament production process. Future work might address the application of DOE to printing parameters and also extend to the adaptation of these to different types of material. The effects of temperature and speed over the long term on material properties can be examined in relation to the stress-relaxation and fatigue resistance.\u003c/p\u003e\n\u003cp\u003eThe analyses were done using machine learning and ensemble methods with five different models. Among them, the best validation accuracy of 99.9% was from the Extra Trees algorithm. This algorithm outperformed others in both accuracy and error rates.\u003c/p\u003e\n\u003cp\u003eFurther research may serve to test other types of thermoplastic materials in light of the findings presented in this work, together with the possibility of taking into account the effects on specimens with more complicated geometrical shapes, which may be of greater value for the possible use in industry.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSelim Bacak: Conceptualization; methodology; software; formal analysis; resources; writing \u0026ndash; original draft; writing \u0026ndash; review and editing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author declare that there is no conflict of interest in the preparation and publication of this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study did not receive any external funding support.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSheoran, A. J., \u0026amp; Kumar, H. (2020). Optimization of FDM process parameters. Materials Today: Proceedings, 21, 1659\u0026ndash;1672.\u003c/li\u003e\n\u003cli\u003eJandyal, A., et al. (2022). Review of processes, materials, and applications in FFF. Sustainable Operations and Computers, 3, 33\u0026ndash;42.\u003c/li\u003e\n\u003cli\u003eFontana, L., et al. (2022). Influence of 3D printing parameters on tensile strength. Materials Today: Proceedings, 57, 657\u0026ndash;663.\u003c/li\u003e\n\u003cli\u003eI. Gibson, D. Rosen, B. Stucker, M. Khorasani, Development of additive manufacturing technology, in: Additive Manufacturing Technologies, Springer, Cham, 2021, https://doi.org/10.1007/978-3-030-56127-7_2..\u003c/li\u003e\n\u003cli\u003eS. Kumar, I. Singh, S.S.R. Koloor, D. Kumar, M.Y. Yahya, On laminated object manufactured FDM-printed ABS/TPU multimaterial specimens: an insight into mechanical and morphological characteristics, Polymers 14 (19) (2022) 4066,\u003c/li\u003e\n\u003cli\u003eA.R.T. Perez, D.A. Roberson, R.B. Wicker, Fracture surface analysis of 3D-Printed tensile specimens of novel ABS-Based materials, J. Fail. Anal. Prev. 14 (3) (2014) 343\u0026ndash;353, https://doi.org/10.1007/s11668-014-9803-9.\u003c/li\u003e\n\u003cli\u003eChac\u0026oacute;n, J. M., et al. (2017). Additive manufacturing of ABS: Mechanical properties and optimization. Material Design, 124, 143\u0026ndash;157.\u003c/li\u003e\n\u003cli\u003eZou, R., et al. (2016). Elasticity and mechanical response of 3D printed materials. Composites Part B, 99, 506\u0026ndash;513.\u003c/li\u003e\n\u003cli\u003eSubramaniam, S. R., et al. (2019). Evaluating ABS tensile strength with varying parameters. AIP Conference Proceedings, 2059.\u003c/li\u003e\n\u003cli\u003eBirosz, M\u0026aacute;rton Tam\u0026aacute;s and And\u0026oacute;, M\u0026aacute;ty\u0026aacute;s and Ledeny\u0026aacute;k, D\u0026aacute;niel, Effect of FDM Infill Patterns on Mechanical Properties. Available at SSRN: https://ssrn.com/abstract=3950131 or http://dx.doi.org/10.2139/ssrn.3950131\u003c/li\u003e\n\u003cli\u003eAhmed, Syed Waqar, Ghulam Hussain, Khurram Altaf, Sadaqat Ali, Mohammed Alkahtani, Mustufa Haider Abidi, and Ayoub Alzabidi. On the effects of process parameters and optimization of interlaminate bond strength in 3D printed ABS/CF-PLA composite. Polymers, 12, 9, (2020) 2155\u003c/li\u003e\n\u003cli\u003eNatureWorks. (2018). ABS technical data sheet. Retrieved from https://www.natureworksllc.com.\u003c/li\u003e\n\u003cli\u003eChen, K., et al. (2021). Optimization of printing parameters for ABS composites. Thin-Walled Structures, 164, 107717.\u003c/li\u003e\n\u003cli\u003eM. Mirzaee, S. Noghanian, L. Wiest, I. Chang, July). Developing flexible 3D printed antenna using conductive ABS materials, in: 2015 IEEE International Symposium on Antennas and Propagation \u0026amp; USNC/URSI National Radio Science Meeting, 2015\u003c/li\u003e\n\u003cli\u003eS. Ahn, M. Montero, D. Odell, S. Roundy, P.K. Wright, Anisotropic material properties of fused deposition modeling ABS, Rapid Prototyp. J. 8 (4) (2002) 248\u0026ndash;257, https://doi.org/10.1108/13552540210441166.\u003c/li\u003e\n\u003cli\u003eTich\u0026yacute;, T.; \u0026Scaron;efl, O.; Vesel\u0026yacute;, P.; Du\u0026scaron;ek, K.; Bu\u0026scaron;ek, D. Mathematical modelling of temperature distribution in selected parts of fff printer during 3d printing process. Polymers 2021, 13, 4213. \u003c/li\u003e\n\u003cli\u003eRamos, N.; Mittermeier, C.; Kiendl, J. Experimental and numerical investigations on heat transfer in fused filament fabrication 3D-printed specimens. Int. J. Adv. Manuf. Technol. 2022, 118, 1367\u0026ndash;1381. \u003c/li\u003e\n\u003cli\u003ePhan, D.D.; Horner, J.S.; Swain, Z.R.; Beris, A.N.; Mackay, M.E. Computational fluid dynamics simulation of the melting process in the fused filament fabrication additive manufacturing technique. Addit. Manuf. 2020, 33, 101161.\u003c/li\u003e\n\u003cli\u003eZhu, Q.; Liu, Z.; Yan, J. Machine learning for metal additive manufacturing: Predicting temperature and melt pool fluid dynamics using physics-informed neural networks. Comput. Mech. 2021, 67, 619\u0026ndash;635.\u003c/li\u003e\n\u003cli\u003eM. Galetto, E. Verna, G. Genta, Effect of process parameters on parts quality and process efficiency of fused deposition modeling, Computer and Industrial engineering 156 (2021) 107238, https://doi.org/10.1016/j.cie.2021.107238.\u003c/li\u003e\n\u003cli\u003eZ. Weng, J. Wang, T. Senthil, L. Wu, Mechanical and thermal properties of ABS/ montmorillonite nanocomposites for fused deposition modeling 3D printing, Mater. Des. 102 (2016) 276\u0026ndash;283, https://doi.org/10.1016/j. matdes.2016.04.045.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"3D Printing, ABS, Tensile Strength, Machine Learning, Ensemble Methods","lastPublishedDoi":"10.21203/rs.3.rs-8605121/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8605121/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Fused Filament Fabrication benefits from Akrilonitril Butadien Stiren due to its mechanical and thermal resistance, yet optimizing extrusion parameters for improved tensile strength and dimensional accuracy remains challenging. This study aims to analyze the impact of extrusion parameters on the tensile resistance and dimensional consistency of Akrilonitril Butadien Stiren filaments using machine learning methodologies to enhance 3D printing efficiency. Akrilonitril Butadien Stiren granules were extruded via a single-screw extruder at 225°C–245°C and screw speeds of 2–6 rpm, with the Taguchi method optimizing the experimental design. Machine learning models, including ensemble methods, were developed to predict tensile strength and filament diameter, with the Extra Trees algorithm achieving the highest accuracy (~99%). The optimal tensile strength (28.5 MPa) and filament diameter (±1.75 mm) were obtained at 235°C and 4 rpm, with temperature (81.6%) having a greater effect on tensile strength than screw speed (18.4%). This study demonstrates the successful integration of classical optimization and machine learning techniques, providing a framework for enhancing Akrilonitril Butadien Stiren filament production quality in 3D printing.","manuscriptTitle":"The Effect of Variable Extrusion Parameters on the Tensile Strength and Production Quality of ABS Filaments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-20 10:48:08","doi":"10.21203/rs.3.rs-8605121/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8e0075d0-603d-4bb7-83b4-f93319d4bd40","owner":[],"postedDate":"January 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-27T12:54:13+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-20 10:48:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8605121","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8605121","identity":"rs-8605121","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}