Prediction and Optimization of Surface quality and Microhardness using Machine learning in Selective laser melting of SS316L Biomedical alloy

Prediction and Optimization of Surface quality and Microhardness using Machine learning in Selective laser melting of SS316L Biomedical alloy | 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 Prediction and Optimization of Surface quality and Microhardness using Machine learning in Selective laser melting of SS316L Biomedical alloy Amit Sharma, Tauseef Uddin Siddiqui, Manoj Kumar Singh, Arshad Noor Siddiquee, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7302973/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 Laser additive manufacturing based selective laser melting (SLM) technique have attracted a lot of attention due to the rising need of high performance materials in aerospace, automotive, and biomedical applications. However, because of the intricate relationships between several parameters including laser power, scanning speed, hatch spacing, and layer thickness, optimization of the process parameters for SLM is a tedious task. Machine learning (ML) technique can handle a variety of data sets and can accurately predict complicated and non-linear relationships in SLM. In this paper, three tree based ML models such as Random forest, Gradient Boosting, and XG boost Regressor are used for prediction of surface roughness (R a ) and microhardness (MH) of SLM fabricated parts for improved part quality and longevity. The efficacy of the ML models is evaluated in terms of prediction accuracy and computational efficiency after training and testing to predict optimal process parameters for minimum R a and maximum MH, respectively. The average error of XG boost model for prediction of R a and MH is 0.1217% and 1.73%, respectively which is significantly lower as compared to Random forest and Gradient boosting methods. Therefore, XG boosting showed better accuracy in prediction of R a and MH values as compared to Random forest and Gradient boosting methods. This is because of its better data handling capacity and efficient capturing of complex data sets. A 29.64% decrease in R a and 14.73% increase in MH values are achieved at optimized settings for performance improvement of SLM fabricated parts. The maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different energy densities after image processing by Image J software. This work will be useful in implementation of ML technique in SLM fabrication for better process control, reduction in trial-and-error, and to improve the functionality and reliability of finished parts. Selective laser melting SS316L alloy Surface roughness Microhardness Machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 1 Introduction The fourth industrial revolution, or Industry 4.0, places a strong emphasis on automation, data-driven programming, and contemporary production methods. It highlights the challenge of achieving creative flexibility while building intricate structures that are valuable[ 1 , 2 ]. A key technology in the present period, additive manufacturing (AM), often known as 3D printing, enables layer-by-layer product fabrication and digital manufacturing techniques. By linking, hardening, or depositing components, AM is a cutting-edge technique for converting CAD models into actual fabricated parts. Metal additive manufacturing (MAM) creates complex component designs by layering and melting metal powders using a high intensity beam [ 3 ]. MAM methods can be divided into groups according on the energy source, such as laser beams or electrons. Selective Laser Melting (SLM), one of these laser-based MAMs, has achieved enormous traction in major industries including aerospace [ 4 ], medicinal [ 5 ], and automotive [ 6 ]. MAM technology has made it possible to create complex components made of different types of materials. Steel is dominantly utilized material in SLM fabrication. Austenitic and low-carbon, AISI316L (SS316L) stainless steel (SS) prints exceptionally well in MAM [ 7 ]. SS 316L is widely used in the medical field (implants, surgical equipment, etc.) due to its excellent biocompatibility; in the chemical, pharmaceutical, and marine industries because of its robust resistance to corrosion; and in the automotive and aerospace industries because of its suitable mechanical properties [ 8 – 10 ]. Tucho and colleagues [ 11 ]examined the microstructure and defects of SS316L parts. It was observed complexity and indications of martensite or ferrite phases. The density of SLM fabricated 316L stainless steel increased, whereas microhardness and corrosion resistance decreased, according to Lin et al.[ 12 ] findings. Huang et al. [ 13 ] discovered that energy density is the primary factor influencing the mechanical properties and density of SLMed components. They also discovered that an increase in hatch spacing (HS) led to the formation of voids, which reduces the density. According to research by Deng et al.[ 14 ], the relative density and Ra of SLM fabricated parts are significantly impacted by laser power (LP) and scanning speed (SS), whereas HS has a minor effect. According to Larimian et al. [ 15 ], scanning technique, SS, and energy density all have an impact on the mechanical properties, microstructure, and densification of 316L alloy under study. Cherry et. al.[ 16 ] revealed that material hardness rises with VED up to 225 HV at 125 J/mm 3 , whereas further rise resulting in a decrease in hardness. Yusuf et. al.[ 17 ] discovered that the mean micro hardness of SLM samples in the x-y, x-z, and y-z planes are 262 HV, 237 HV, and 239 HV, correspondingly. The maximum micro hardness was found in the x-y plane, demonstrating anisotropy in SLM, which is characteristic of AM methods for metal elements. This is owing to the layer-wise construction method combined with the "island" scan strategy, leading to heterogeneous morphologies and asymmetric grain formations. Pagac et. al. [ 18 ] concluded that heat treatment and laser performance have no major influence on microhardness of the SLM fabricated components. SLM fabricated SS 316 L specimens is around 50 HBW harder, possibly owing to grain size, chemical improvements as well as internal tension. These parameters influence the component's hardness, according to the mean grain size in rolled steel 316L. Manikandan et. al.[ 19 ] discovered that micro hardness values grew along with rising VED, with a larger association reported for surface PPBD than surface PBD. Surface PPBD had the greatest microhardness value at 69 J/mm³. Surface PBD had the highest at 308.8 ± 8.7 HV and the lowest at 292.7 ± 15.7 HV. Barrionuevo et. al. [ 20 ] discovered a direct correlation among microhardness and relative density, with higher density leading in increased micro hardness. The mean micro hardness found was approximately 237 ± 9 HV. Nevertheless, outcomes varied, with the only evident connection being power and speed, with more power and lower speed leading in greater micro hardness. A rough surface appearance caused by laser energy input, powder particle size, and material properties is the result of defect formation in SLMed parts, including lack of fusion, residual unmelted powder, open pores, material stacking, and balling effects. These defects may result in interlayer and multilayer defects [ 21 – 23 ]. LP and SS are key factors that influence the Ra of SLM-fabricated parts, which are influenced by track width, scan space, and powder layer thickness [ 24 ]. Surface quality may suffer from an excessive laser energy density. The R a of the component is affected by the wiper motion, gas flow, and scan direction [ 25 ]. In LPBF process, R a value frequently surpasses 10 µm, while in mechanical processes like milling and grinding, it is between 1–2 µm. Because this roughness serves as a site for crack formation, AM components have low fatigue limits [ 26 ]. Surface improvement parameters are frequently adjusted in the study conducted by Galy et al. [ 27 ] according to the laser energy input. Machine learning (ML) enables machines to learn and perform operations autonomously. ML is categorized into supervised learning, unsupervised learning, and reinforcement learning [ 28 ]. ML techniques enable effective data-driven strategies for maximizing the quality of SLM components by process parameter optimization, microstructure prediction and defect detection [ 29 – 31 ]. However, the efficiency of these techniques relies on adequacy of collected data through in-situ sensors, feature extraction and filtering of data collected [ 32 , 33 ]. Enhancing the efficiency and excellence of certain components is a vital research goal in the manufacturing process using the SLM technology. Moreover, the integration of Artificial Intelligence (AI) and ML algorithms may enhance the optimization of process signatures in real-time, resulting in accelerated and more effective production in L-PBF. An ensemble ML technique called Random Forest (RF) constructs several decision trees during training and aggregates their predictions to increase accuracy and decrease over-fitting. Gradient Boosting (GB), a progressive ML technique, improves classification and regression prediction accuracy by constructing trees to repair prior errors [ 34 ]. Extreme Gradient Boosting (XG boost) is a well-liked and successful use of the GB technique for regression predictive modeling. It is often utilized in ML modeling because of its great performance and fast computation speed [ 35 ]. In order to anticipate and optimize process parameters, numerous studies have been carried out on the SLM technology and ML utilizing various materials [ 36 – 42 ]. Prediction and optimization of process parameters during SLM fabrication of SS316L alloy utilizing tree-based boosting ML models has not been extensively studied till now. These boosting strategies use ensemble learning to construct a strong classifier from several weak classifiers for accurate prediction of R a and MH values. This study uses training and testing data sets to predict R a and MH values using Random Forest, Gradient Boosting, and XG Boosting techniques. The optimal ML model for maximizing the performance of SLM fabricated parts was determined by comparing experimental and predicted values. For detection and characterization of surface defects, SEM is used to capture images under different process settings. Image J software is also utilized for image processing and analysis of the SEM images for porosity mapping and quantification at different energy densities during SLM fabrication. 2 Materials and Methods 2.1 SS 316L alloy For the sample preparation, powdered austenitic stainless steel AISI 316L alloy with a particle size of 15–60 µm was utilized. The chrome-nickel austenitic alloy is known as SS316L. Additionally, it contains molybdenum, which improves resistance to pitting from chloride ion solutions, boosts strength at high temperatures, and increases general corrosion resistance. Additionally, alloy 316L offers superior formability and fabricability. This alloy used in medical implants, architecture items, heat exchangers, fasteners, marine use, and in food equipment etc[ 43 ]. The chemical composition of the SS316L powder is shown in Table 1 . Table 1 Chemical composition of SS316L powder Elem. Fe Cr Ni Mo Mn Si O C Cu P S Wt % Bal. 17.7 12.7 2.36 0.65 0.62 0.03 0.022 0.02 0.007 0.005 2.2 Experimental procedure The SLM 280 is a selective laser melting machine utilized in this work for fabrication of components made of SS316L alloy. It has a beam diameter of 80 µm, a fiber laser type, and a power of 400W. This SLM machine was employed to fabricate cylindrical samples, as seen in Fig. 1 , with a diameter of 10 mm and a height of 5.0 mm. A CAD model of required shape and size has been prepared and exported to the SLM machine for fabrication using SS316L alloy powder (particle size: 30 µm). The LP, SS and HS are used as variable parameters while other parameters were kept as constant. All the experiments have been conducted using L 27 Taguchi design of experiment within the range of selected parameters. Table 2 displays the range of process parameters that were chosen for performing SLM fabrication of SS316L alloy. Table 2 Range of process parameters S No Parameters Range 1 2 3 1 Laser power (LP) 225 275 325 2 Scanning speed (SS) 650 750 850 3 Hatch spacing (HS) 0.10 0.11 0.12 2.3 Measurement of surface roughness The unevenness of a machined surface, which is defined by tiny peaks and valleys that are closely spaced, is referred to as the R a in machining. It plays a crucial role in forecasting the surface quality of components since it has a big impact on how well the parts function during fatigue loading. Durability and performance are adversely affected by surface asperities. As seen in Fig. 2 , the R a of the manufactured samples was determined using a surface roughness tester [Make: Mitutoyo Japan; Make: SJ-201]. To increase measurement precision and decrease error, each observation was taken thrice, and the average value was calculated. 2.4 Measurement of microhardness Microhardness (MH) is a material's hardness tested by forcing an indenter, such as a Vickers or Knoop indenter, into its surface under a stress ranging from 15 to 1000 gf; the indentations are often so minute that they must be determined with a microscope. The resin mold samples of required dimensions have been prepared for testing purposes. The Micro Wizhard Microhardness testing machine Ver.1.401 (Mitutoyo: Japan) is used for measurement of MH of SLM fabricated samples. The 980 gf indentation load has been applied for a dwell period of 15 seconds and using grid pattern as shown in Fig. 3 . 2.5 Machine learning models In the present work, three ensemble ML models have been utilized for prediction of R a and MH values and performance comparison at different process settings. Jupyter NoteBook and Python scripting were used for preprocessing and data modeling of three learning algorithms as shown in Table 3 . The experimental data for R a and MH were collected and utilized to train three ML algorithms such as Random forest (RF), Gradient booting (GB) and XG boosting (XGB) to simulate the relationship between the input parameters and output response to reliably forecast R a and MH values for unknown input parameters like LP, SS and HS. MAE, MSE, RMSE and MPE are effectiveness measures for ML models. The details of RF, GB, and XG boost ML model parameters are shown in Table 4 . Table 3 Different libraries used in Python S No Name of library 1 Pandas 2 sklearn.ensemble.GradientBoostingRegressor 3 sklearn.ensemble.AdaBoostRegressor 4 xgboost.XGBRegressor 5 sklearn.metrics 6 Numpy 7 matplotlib.pyplot Table 4 Details of RF, GBM and XG boost model parameters Parameters/Methods RF GBM XG boost n_estimators 50 100 100 max_depth --- 3 6 Learning rate 1.0 0.1 0.1 random_state 42 42 42 2.6 Evaluation of Machine learning models In this study, statistical tests were performed to assess the prediction accuracy of the developed ML models. These measures include MSE, MAE and RMSE. These metrics are calculated using the equations below. $$\:MSE=\frac{1}{N}\underset{i=1}{\overset{N}{?}}{e}_{i}^{2}$$ 1 $$\:MAE=\frac{1}{N}\underset{i=1}{\overset{N}{?}}\left|{e}_{i}\right|$$ 2 $$\:RMSE=\sqrt{\frac{1}{N}\underset{i=1}{\overset{N}{?}}{e}_{i}^{2}}$$ 3 In this case, N is the number of testing data samples, and e i is the error difference between the experimental and predicted R a and MH values for each data set. MSE, MAE, and RMSE are metrics used to monitor and evaluate the predicted models performance and accuracy. 3 Results and Discussion 3.1 XRD Analysis X-ray diffraction (2D-XRD) was performed using a Bruker D8 ADVANCE. The samples were examined using a Cu K/α radiation (λ = 1.54 Å) source at a voltage of 40 kV and current of 15 mA. The scan speed was 6 degrees/minute throughout a range of 5 o to 100 o . It also shows the existence of a single austenitic FCC phase, with no foreign phase peaks detected. The crystallographic planes {111}, {200}, {220}, and {311} were at 2θ angles of 43.60°, 50.759°, 74.595°, and 90.487°[ 44 ], correspondingly as shown in Fig. 4 . 3.2 Comparison of experimental and predicted surface roughness The comparison of experimental and predicted values of R a for training and testing data sets is shown in Fig. 5 . Based on the graph, Table 5 shows results of different ML Models and Error Metrics in terms of MAE, MSE and RMSE. It was found that values of MAE, MSE and RMSE for XG boosting were 0.73, 1.37 and 1.21 which are significantly lower values as compared to RF and GB methods. XG boosting showed 29.51% and 19.88% better accuracy in prediction of R a values as compared to RF and GB methods, respectively. The average error between experimental and predicted values of R a for RF and GB methods for the training and testing data sets were 2.45% and 2.16% respectively as shown in Fig. 6 and Fig. 7 . In contrast, the respective values for XG boost model were 1.73%, respectively which is considerably lower value as compared to RF and GB model as shown in Fig. 8 . This is because of its better data handling capacity and efficient capturing of complex data sets [ 35 , 45 ]. The minimum value of R a were found to be 9.41µm at optimum parameter settings i.e. 225W LP, 850 mm/s SS and 0.10 mm HS as shown by S/N ratio plot for R a in Fig. 9 . A 29.64% reduction in R a value is achieved at optimum parameter settings as compared to worst setting. Table 5 Comparison of ML Models and Error Metrics for R a Model MAE MSE RMSE MPE Random forest 0.98 1.48 1.97 1.83 Gradient Boosting 0.96 1.42 1.68 1.61 XG Boosting 0.73 1.37 1.21 1.29 3.3 Comparison of experimental and predicted microhardness The comparison of experimental and predicted values of MH for training and testing data sets is shown in Fig. 10 . Based on these graph, Table 6 shows comparison of different ML Models and Error Metrics in terms of MAE, MSE and RMSE. It was found that values of MAE, MSE and RMSE for XG boosting were 0.6687, 1.4478 and 1.2032 which are significantly lower values as compared to RF and GB methods. XG boosting showed 38.50% and 23.07% better accuracy in prediction of MH values as compared to RF and GBM methods, respectively. The average error between experimental and predicted values of MH for RF and GB methods for the training and testing data sets were 0.6465% and 0.3519% respectively as shown in Fig. 11 and Fig. 12 . In contrast, the respective values for XG boost model were 0.1217%, which is a considerably lower value as compared to RF and GB model as shown in Fig. 13 . This is because of its better data handling capacity and efficient capturing of complex data sets [ 35 , 45 ]. The maximum value of MH was found to be 241.80 HV at optimum parameter settings i.e. 275W LP, 650 mm/s SS and 0.10 mm HS as shown by S/N ratio plot for MH in Fig. 14 . A 14.73% improvement in microhardness is achieved at optimum parameter settings as compared to worst setting. Table 6 Comparison of ML Models and Error Metrics for MH Model MAE MSE RMSE MPE Random Forest 1.0642 1.8028 1.3427 0.26 Gradient Boosting 0.8296 1.4920 1.2802 0.19 XG Boosting 0.6687 1.4478 1.2032 0.12 3.4 Feature correlation heatmap In ML technique, feature correlation heatmap is an effective visual aid for comprehending the connections among various features in a dataset. Each cell's color indicates the strength and direction of the connection, and it shows the correlation coefficients between every pair of features in a matrix. Figure 15 (a) shows that LP has a strong linear relationship with R a , while SS had a negative correlation with R a , demonstrating an inverse relationship. Figure 15 (b) shows that SS and HS had a strong negative correlation with MH, demonstrating an inverse relationship. While LP has no significant relationship with MH. 3.5 Analysis of variance The analysis of variance (ANOVA) has been conducted to assess variation within and between groups for R a for smaller-the-better S/N ratios and for MH for larger-the-better, as shown in Table 7 . Based on the p-values (for 95% confidence interval), LP has the most dominant effect on R a as shown in Table 7 . Lower LP, high SS and HS spacing is required to achieve minimum value of R a . At the optimum setting, R a is found to be 9.705 µm which is 29.60.5% lower than its value at worst parameter settings (13.832 µm). Moderate LP, lower SS and lower HS is required to achieve maximum value of MH. At the optimum parameter settings, MH is found to be 241.80 HV which is 14.73% higher than its value at worst parameter settings (207.37 HV). Medium LP, lower SS and lower HS is required to achieve minimum value of porosity. Based on the p-values (for 95% confidence interval), SS has the most dominant effect on MH value as shown in Table 9 . 3.6 Ranking of process parameters The ranking of each factor is based on delta statistics that compare the relative magnitude of the effects, as indicated in Tables 8 and 10 , respectively. This includes the average of S/N ratio of each feature at each factor level. The delta value is obtained by subtracting the lowest average value of each factor from its greatest average value. The LP (delta value: 2.15) has the most significant effect on the R a followed by SS (delta value: 0.63) and HS (delta value: 0.12) [ 14 , 36 ]. According to the S/N ratio values, SS (delta value: 0.45) has the largest impact on MH, subsequent to HS (delta value: 0.37). LP has the least impact on the value of MH of the SLM fabricated components (delta value: 0.05). The SS has the most dominant effect on MH values followed by HS and LP. Table 7 ANOVA table for R a Source DF SS MS F P Laser power 2 43.2920 21.6460 51.31 0.000 Scanning speed 2 2.8714 1.4357 3.40 0.053 Hatch spacing 2 0.1154 0.0577 0.14 0.873 Error 20 8.4366 0.4218 Total 26 54.7153 R-sq: 95.84, R-sq (adj): 94.77 (for 95% CI) Table 8 Statistical ranking table for R a Level Laser power Scanning speed Hatch spacing 1 -20.46 -22.12 -21.72 2 -22.24 -21.71 -21.75 3 -22.61 -21.49 -21.84 Delta 2.15 0.63 0.12 Rank 1 2 3 Table 9 ANOVA table for MH Source DF SS MS F P Laser power 2 8.47 4.233 2.50 0.107 Scanning speed 2 618.30 309.150 182.55 0.000 Hatch spacing 2 387.06 193.532 114.28 0.000 Error 20 33.87 1.694 Total 26 1047.70 R-sq: 96.77, R-sq (adj): 95.80 (for 95% CI) Table 10 Statistical ranking table for MH Level Laser power Scanning speed Hatch spacing 1 46.74 46.99 46.92 2 46.75 46.67 46.73 3 46.70 46.54 46.55 Delta 0.05 0.45 0.37 Rank 3 1 2 3.7 Surface defects and porosity mapping Various surface defects in SLM fabricated parts are find out using SEM (Make: Jeol, Model: JSM6610LV) at different process settings as depicted in Fig. 16 . Keyhole porosity, binding porosity, lack of fusion, balling, cracking, and other surface imperfections can significantly affect the mechanical properties and performance of SLM fabricated parts. It is possible that trapped gases in the molten pool are the cause of gas porosity. When there is insufficient melting and bonding within a layer or between powder layers, there is no fusion. Inadequate scanning or insufficient energy intake may be responsible for occurrence of such types of defects during SLM fabrication. Keyhole porosity, which results in the formation of a gas cavity during melting, can occur when excessive laser energy is supplied during SLM fabrication of SS316 alloy. These SEM images were further treated by Image J software for digital image processing and porosity mapping as shown in Fig. 17 (a). This is open source software operating on Java platform for processing and analyzing scientific images [imagej.net]. The maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different volume energy densities (VEDs) after image processing by image J software as shown in Fig. 17 (b). 4. Conclusions In this work, three tree based ML models such as Random forest, Gradient boosting and XG boosting are used for prediction and optimization of R a and MH values of SLM fabricated parts made of SS316L alloy. Following conclusions have been obtained from the present research: These ensemble learning ML models can provide better accuracy and reliability for predictive modeling for R a and MH under different process settings. The MAE, MSE, and RMSE statistical values obtained from testing data are found to be 0.73, 1.37, and 1.21, for R a values. The MAE, MSE, and RMSE statistical values obtained from testing data are found to be 0.6687, 1.4478, and 1.2032, for MH values. The testing process revealed that the closest result to the experimental values was delivered by the XG booting model. The average error of XG boost model is 0.1217% and 1.73%, respectively which is significantly lower as compared to RF and GB methods. Therefore, XG boosting showed better accuracy in prediction of R a and MH values as compared to Random forest and Gradient boosting methods. XG boosting showed 29.51% and 19.88% better accuracy in prediction of R a values as compared to RF and GB methods. XG boosting showed 38.50% and 23.07% better accuracy in prediction of MH values as compared to RF and GBM methods. Hence, XG boosting can be adopted for the prediction of R a and MH values of SLM fabricated samples made of SS316L powder. A 29.64% reduction in R a value is obtained at optimum process settings as compared to worst setting. This will improve the performance and durability of fabricated parts. A 14.73% improvement in MH value is obtained at optimum process settings as compared to worst setting. The maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different energy densities by image processing by image J software. Most of the pores were found spherical shaped due to gas atomization of SS316L powder. Using ML models in metal additive manufacturing will help enhance process control, decrease trial and error, and increase the quality and longevity of the final SLM fabricated parts. Declarations Competing Interest The authors declare no competing interests. Ethical Approval Ethical Approval declaration is not applicable. Funding No funding was received for this work. Author contributions A.S.: conceptualization, formal analysis, methodology, investigation, resources and writing—original draft; T.U.S: formal analysis, investigation, writing—review and editing, supervision and validation; M.K.S.: investigation, writing—review and editing, supervision and validation; A.N.S.: supervision, methodology, writing—review and editing, and validation; T.B.: methodology, formal analysis, investigation, resources and writing—original draft Availability of data and materials The data of this study are available from the corresponding author upon reasonable request. References Butt, J.: Exploring the Interrelationship between Additive Manufacturing and Industry 4.0. Designs. 4 , 13 (2020). https://doi.org/10.3390/designs4020013 Dilberoglu, U.M., Gharehpapagh, B., Yaman, U., Dolen, M.: The Role of Additive Manufacturing in the Era of Industry 4.0. Procedia Manuf. 11 , 545–554 (2017). https://doi.org/10.1016/j.promfg.2017.07.148 Aversa, A., Saboori, A., Marchese, G., et al.: Recent Progress in Beam-Based Metal Additive Manufacturing from a Materials Perspective: A Review of Patents. J. Materi Eng. Perform. 30 , 8689–8699 (2021). https://doi.org/10.1007/s11665-021-06273-3 Stolt, R., Elgh, F.: Introducing design for selective laser melting in aerospace industry☆. J. Comput. Des. Eng. 7 , 489–497 (2020). https://doi.org/10.1093/jcde/qwaa042 Talib Mohammed, M.: Mechanical Properties of SLM-Titanium Materials for Biomedical Applications: A Review. Materials Today: Proceedings 5:17906–17913. (2018). https://doi.org/10.1016/j.matpr.2018.06.119 Yang, J., Li, B., Liu, J., et al.: Application of Additive Manufacturing in the Automobile Industry: A Mini Review. Processes. 12 , 1101 (2024). https://doi.org/10.3390/pr12061101 Xu, M., Guo, H., Wang, Y., et al.: Mechanical properties and microstructural characteristics of 316L stainless steel fabricated by laser powder bed fusion and binder jetting. J. Mater. Res. Technol. 24 , 4427–4439 (2023). https://doi.org/10.1016/j.jmrt.2023.04.069 Li, N., Huang, S., Zhang, G., et al.: Progress in additive manufacturing on new materials: A review. J. Mater. Sci. Technol. 35 , 242–269 (2019). https://doi.org/10.1016/j.jmst.2018.09.002 Peng, T., Chen, C.: Influence of energy density on energy demand and porosity of 316L stainless steel fabricated by selective laser melting. Int. J. Precis Eng. Manuf-Green Tech. 5 , 55–62 (2018). https://doi.org/10.1007/s40684-018-0006-9 Yap, C.Y., Chua, C.K., Dong, Z.L., et al.: Review of selective laser melting: Materials and applications. Appl. Phys. Reviews. 2 , 041101 (2015). https://doi.org/10.1063/1.4935926 Tucho, W.M., Lysne, V.H., Austbø, H., et al.: Investigation of effects of process parameters on microstructure and hardness of SLM manufactured SS316L. J. Alloys Compd. 740 , 910–925 (2018). https://doi.org/10.1016/j.jallcom.2018.01.098 Lin, K., Gu, D., Xi, L., et al.: Selective laser melting processing of 316L stainless steel: effect of microstructural differences along building direction on corrosion behavior. Int. J. Adv. Manuf. Technol. 104 , 2669–2679 (2019). https://doi.org/10.1007/s00170-019-04136-9 Huang, M., Zhang, Z., Chen, P.: Effect of selective laser melting process parameters on microstructure and mechanical properties of 316L stainless steel helical micro-diameter spring. Int. J. Adv. Manuf. Technol. 104 , 2117–2131 (2019). https://doi.org/10.1007/s00170-019-03928-3 Deng, Y., Mao, Z., Yang, N., et al.: Collaborative Optimization of Density and Surface Roughness of 316L Stainless Steel in Selective Laser Melting. Materials. 13 , 1601 (2020). https://doi.org/10.3390/ma13071601 Larimian, T., Kannan, M., Grzesiak, D., et al.: Effect of energy density and scanning strategy on densification, microstructure and mechanical properties of 316L stainless steel processed via selective laser melting. Mater. Sci. Engineering: A. 770 , 138455 (2020). https://doi.org/10.1016/j.msea.2019.138455 Cherry, J.A., Davies, H.M., Mehmood, S., et al.: Investigation into the effect of process parameters on microstructural and physical properties of 316L stainless steel parts by selective laser melting. Int. J. Adv. Manuf. Technol. 76 , 869–879 (2015). https://doi.org/10.1007/s00170-014-6297-2 Yusuf, S., Chen, Y., Boardman, R., et al.: Investigation on Porosity and Microhardness of 316L Stainless Steel Fabricated by Selective Laser Melting. Metals. 7 , 64 (2017). https://doi.org/10.3390/met7020064 Pagáč, M., Hajnyš, J., Petrů, J., Zlámal, T.: Comparison of Hardness of Surface 316L Stainless Steel Made by Additive Technology and Cold Rolling. MSF. 919 , 84–91 (2018). https://doi.org/10.4028/www.scientific.net/MSF.919.84 Manikandan, P., Venkatesan, K.: Role of volumetric energy density on surface quality and mechanical properties of selective laser melted 310 stainless steel. Results Eng. 25 , 104479 (2025). https://doi.org/10.1016/j.rineng.2025.104479 Barrionuevo, G.O., Ramos-Grez, J.A., Sánchez-Sánchez, X., et al.: Influence of the Processing Parameters on the Microstructure and Mechanical Properties of 316L Stainless Steel Fabricated by Laser Powder Bed Fusion. J. Manuf. Mater. Process. 8 , 35 (2024). https://doi.org/10.3390/jmmp8010035 Zhou, X., Wang, D., Liu, X., et al.: 3D-imaging of selective laser melting defects in a Co–Cr–Mo alloy by synchrotron radiation micro-CT. Acta Mater. 1–16 (2015) Yu, W.H., Sing, S.L., Chua, C.K., et al.: Particle-reinforced metal matrix nanocomposites fabricated by selective laser melting: A state of the art review. Prog. Mater. Sci. 104 , 330–379 (2019). https://doi.org/10.1016/j.pmatsci.2019.04.006 Sharma, H., Singla, J., Singh, V., et al.: Influence of post heat treatment on metallurgical, mechanical, and corrosion analysis of wire arc additive manufactured inconel 625. J. Mater. Res. Technol. 27 , 5910–5923 (2023). https://doi.org/10.1016/j.jmrt.2023.11.074 Wang, D., Liu, Y., Yang, Y., Xiao, D.: Theoretical and experimental study on surface roughness of 316L stainless steel metal parts obtained through selective laser melting. RPJ. 22 , 706–716 (2016). https://doi.org/10.1108/RPJ-06-2015-0078 Li, B.-Q., Li, Z., Bai, P., et al.: Research on Surface Roughness of AlSi10Mg Parts Fabricated by Laser Powder Bed Fusion. Metals. 8 , 524 (2018). https://doi.org/10.3390/met8070524 Abd-Elaziem, W., Elkatatny, S., Sebaey, T.A., et al.: Machine learning for advancing laser powder bed fusion of stainless steel. J. Mater. Res. Technol. 30 , 4986–5016 (2024). https://doi.org/10.1016/j.jmrt.2024.04.130 Galy, C., Le Guen, E., Lacoste, E., Arvieu, C.: Main defects observed in aluminum alloy parts produced by SLM: From causes to consequences. Additive Manuf. 22 , 165–175 (2018). https://doi.org/10.1016/j.addma.2018.05.005 Mahmoud, D., Magolon, M., Boer, J., et al.: Applications of machine learning in process monitoring and controls of L-PBF additive manufacturing: A review. Appl. Sci. 11 , 11910 (2021) Eshkabilov, S., Ara, I., Azarmi, F.: A comprehensive investigation on application of machine learning for optimization of process parameters of laser powder bed fusion-processed 316L stainless steel. Int. J. Adv. Manuf. Technol. 123 , 2733–2756 (2022). https://doi.org/10.1007/s00170-022-10331-y Okaro, I.A., Jayasinghe, S., Sutcliffe, C., et al.: Automatic fault detection for laser powder-bed fusion using semi-supervised machine learning. Additive Manuf. 27 , 42–53 (2019). https://doi.org/10.1016/j.addma.2019.01.006 Barrionuevo, G.O., Ramos-Grez, J.A., Walczak, M., Betancourt, C.A.: Comparative evaluation of supervised machine learning algorithms in the prediction of the relative density of 316L stainless steel fabricated by selective laser melting. Int. J. Adv. Manuf. Technol. 113 , 419–433 (2021). https://doi.org/10.1007/s00170-021-06596-4 Caggiano, A., Zhang, J., Alfieri, V., et al.: Machine learning-based image processing for on-line defect recognition in additive manufacturing. CIRP Ann. 68 , 451–454 (2019). https://doi.org/10.1016/j.cirp.2019.03.021 Delli, U., Chang, S.: Automated Process Monitoring in 3D Printing Using Supervised Machine Learning. Procedia Manuf. 26 , 865–870 (2018). https://doi.org/10.1016/j.promfg.2018.07.111 Biau, G., Cadre, B.: Optimization by Gradient Boosting. In: Daouia, A., Ruiz-Gazen, A. (eds.) Advances in Contemporary Statistics and Econometrics: Festschrift in Honor of Christine Thomas-Agnan, pp. 23–44. Springer International Publishing, Cham (2021) Lee, S., Park, J., Kim, N., et al.: Extreme gradient boosting-inspired process optimization algorithm for manufacturing engineering applications. Mater. Design. 226 , 111625 (2023). https://doi.org/10.1016/j.matdes.2023.111625 Aqilah, D.N., Sayuti, A.K.M., Farazila, Y., et al.: Effects of Process Parameters on the Surface Roughness of Stainless Steel 316L Parts Produced by Selective Laser Melting. J. Test. Eval. 46 , 1673–1683 (2018). https://doi.org/10.1520/JTE20170140 Wang, J., Jeong, S.G., Kim, E.S., et al.: Material-agnostic machine learning approach enables high relative density in powder bed fusion products. Nat. Commun. 14 , 6557 (2023). https://doi.org/10.1038/s41467-023-42319-x Tan, X., Chen, D., Xiao, H., et al.: Prediction of phase and tensile properties of selective laser melting manufactured high entropy alloys by machine learning. Mater. Today Commun. 41 , 110209 (2024). https://doi.org/10.1016/j.mtcomm.2024.110209 Toprak, C.B., Dogruer, C.U.: Optimal process parameter determination in selective laser melting via machine learning-guided sequential quadratic programing. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 239:807–820. (2025). https://doi.org/10.1177/09544062241288948 Pashmforoush, F., Seyedzavvar, M.: A transfer learning-based machine learning approach to predict mechanical properties of different material types fabricated by selective laser melting process. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering 09544089231215683. (2023). https://doi.org/10.1177/09544089231215683 Yehia, H.M., Hamada, A., Sebaey, T.A., Abd-Elaziem, W.: Selective Laser Sintering of Polymers: Process Parameters, Machine Learning Approaches, and Future Directions. J. Manuf. Mater. Process. 8 , 197 (2024). https://doi.org/10.3390/jmmp8050197 Chang, L.-K., Jiang, K., Chung, C., et al.: Integrating machine learning and multi-objective optimization to investigate the magnetic and mechanical properties of FeSiCr soft magnetic composite processed by selective laser melting. Int. J. Adv. Manuf. Technol. 132 , 3637–3653 (2024). https://doi.org/10.1007/s00170-024-13589-6 D’Andrea, D.: Additive Manufacturing of AISI 316L Stainless Steel: A Review. Metals. 13 , 1370 (2023). https://doi.org/10.3390/met13081370 Puichaud, A.-H., Flament, C., Chniouel, A., et al.: Microstructure and mechanical properties relationship of additively manufactured 316L stainless steel by selective laser melting. EPJ Nuclear Sci. Technol. 5 , 23 (2019). https://doi.org/10.1051/epjn/2019051 Gadagi, A., Sivaprakash, B., Adake, C., et al.: Epoxy composite reinforced with jute/basalt hybrid – Characterisation and performance evaluation using machine learning techniques. Compos. Part. C: Open. Access. 14 , 100453 (2024). https://doi.org/10.1016/j.jcomc.2024.100453 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-7302973","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":504307088,"identity":"e03eda16-e7c2-45f8-a191-5b72df9df4db","order_by":0,"name":"Amit Sharma","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABC0lEQVRIiWNgGAWjYDACCRiDmYHhgISBjRwDAw/RWpgZH1hUpBmToAWox6DizOHEBkJa5Gc3P/t0o4ZBXred/5jEzba09A3Hzx588IHBTk63AbsWgzvHjGfnHGMw3HaYmU1yZptN7oYzecmGMxiSjc0O4NAikWDMnMPGwAjSIi3Zlpa74UCOmTQPw4HEbTi0yM9I/8yc84/BHqzlb9vhdIPzb/BrYbiRY8yc28aQCNTCbCBx5nCCwQ0CtgAVFDPn9jEkA7UYPpCoSDOceeONseEMA9x+ATpsM3PONwbbbecPPgBFpTzf+RzDBx8q7ORwaYGC/wimAlilAV7l6PY2kKJ6FIyCUTAKRgIAAITIXixotq8zAAAAAElFTkSuQmCC","orcid":"","institution":"M.J.P. Rohilkhand University","correspondingAuthor":true,"prefix":"","firstName":"Amit","middleName":"","lastName":"Sharma","suffix":""},{"id":504307089,"identity":"2dba3a68-e99b-45d5-a08e-1c1b9375852d","order_by":1,"name":"Tauseef Uddin Siddiqui","email":"","orcid":"","institution":"M.J.P. Rohilkhand University","correspondingAuthor":false,"prefix":"","firstName":"Tauseef","middleName":"Uddin","lastName":"Siddiqui","suffix":""},{"id":504307090,"identity":"c0ca3303-05a5-4b80-a8bc-21023d527ff2","order_by":2,"name":"Manoj Kumar Singh","email":"","orcid":"","institution":"M.J.P. Rohilkhand University","correspondingAuthor":false,"prefix":"","firstName":"Manoj","middleName":"Kumar","lastName":"Singh","suffix":""},{"id":504307091,"identity":"0357884a-c0b6-48da-a505-4fc9b6ab40b2","order_by":3,"name":"Arshad Noor Siddiquee","email":"","orcid":"","institution":"Jamia Millia Islamia","correspondingAuthor":false,"prefix":"","firstName":"Arshad","middleName":"Noor","lastName":"Siddiquee","suffix":""},{"id":504307092,"identity":"0d84133b-d1a4-48ac-bcc5-08bf84612f9b","order_by":4,"name":"Tarun Bhardwaj","email":"","orcid":"","institution":"Ajay Kumar Garg Engineering College","correspondingAuthor":false,"prefix":"","firstName":"Tarun","middleName":"","lastName":"Bhardwaj","suffix":""}],"badges":[],"createdAt":"2025-08-05 17:08:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7302973/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7302973/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89979261,"identity":"7eb156b1-12d3-4411-ac2a-bb2cd7f87d01","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":231592,"visible":true,"origin":"","legend":"\u003cp\u003eSLM fabrication of SS316L alloy parts\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/570fed7411dfc96c66050b92.png"},{"id":89981519,"identity":"eef80c04-25d8-46fc-9e6e-ab87350d0807","added_by":"auto","created_at":"2025-08-27 06:25:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":182604,"visible":true,"origin":"","legend":"\u003cp\u003eMeasurement of surface roughness of SLM fabricated parts\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/d256f4c970f641f73d90779b.png"},{"id":89979263,"identity":"34b07ffd-1459-4169-b720-f316f35b2966","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":113738,"visible":true,"origin":"","legend":"\u003cp\u003eMeasurement of microhardness of SLM fabricated parts\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/432206362ab82a114ce4e4b0.png"},{"id":89979264,"identity":"73235b6c-c26c-44be-bf29-a4bd7a6b8df0","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":91911,"visible":true,"origin":"","legend":"\u003cp\u003eXRD plot of SLM fabricated sample at different VED’s\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/2f2daa9b759b0d9194efdf95.png"},{"id":89979269,"identity":"a63b16ed-9bd9-4d74-ae12-05c8df155612","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":87034,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Experimental and predicted R\u003csub\u003ea\u003c/sub\u003e for training and testing data sets\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/faa61c9b0a623fb2b7697ea4.png"},{"id":89981522,"identity":"95c2039d-4460-423e-be38-547a518e9fa3","added_by":"auto","created_at":"2025-08-27 06:25:48","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":49591,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of R\u003csub\u003ea\u003c/sub\u003e by Random forest ML model\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/f10008cd43f3e32063a0af85.png"},{"id":89979271,"identity":"e6917366-dfe8-45e7-b86b-a5327ba16b44","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":48874,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of R\u003csub\u003ea\u003c/sub\u003e by GB ML model\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/7ba7e5787ce366216abef808.png"},{"id":89979267,"identity":"3b9e7f2a-0cc9-4e0f-83ee-ae4da631e780","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":49612,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of R\u003csub\u003ea\u003c/sub\u003e by XG Boost ML model\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/6343649759df1918a80ec02f.png"},{"id":89979280,"identity":"ff453396-a2ec-44f5-b7a7-ff0d69b9b76c","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":32661,"visible":true,"origin":"","legend":"\u003cp\u003eS/N ratio plot for R\u003csub\u003ea\u003c/sub\u003e\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/b1caeafc2b53d51d9abfe52a.png"},{"id":89979278,"identity":"ffa4d44a-9235-4896-9f25-61aed516ed15","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":106171,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Experimental and predicted MH for training and testing data sets\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/4be0578f3835bd43ca909898.png"},{"id":89981525,"identity":"f50a7d7b-cc2a-4c87-ba0a-f0b95ecaa1d9","added_by":"auto","created_at":"2025-08-27 06:25:48","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":67361,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of MH by Random forest ML model\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/f02b7592c4932105b0859abc.png"},{"id":89981535,"identity":"3193f93e-1f0d-4952-9844-dee414231b7d","added_by":"auto","created_at":"2025-08-27 06:25:49","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":68088,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of MH by GB ML model\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/58306bf1c7d2cf613d3c8604.png"},{"id":89981527,"identity":"ee228655-43ba-4e06-9613-91e4daafc31b","added_by":"auto","created_at":"2025-08-27 06:25:49","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":67343,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted and actual values of MH by XG Boost ML model\u003c/p\u003e","description":"","filename":"13.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/e14b8c76312ae64efc769a4d.png"},{"id":89981523,"identity":"cc08a1ff-3205-4ed5-bfd8-4dd06082188a","added_by":"auto","created_at":"2025-08-27 06:25:48","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":46428,"visible":true,"origin":"","legend":"\u003cp\u003eS/N ratio plot for MH\u003c/p\u003e","description":"","filename":"14.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/af84b30963fd2fbb9761af55.png"},{"id":89979282,"identity":"55a7f974-3412-497a-8cf5-a4bcbab54386","added_by":"auto","created_at":"2025-08-27 06:17:48","extension":"png","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":72508,"visible":true,"origin":"","legend":"\u003cp\u003eFeature correlation heatmap for (a) R\u003csub\u003ea\u003c/sub\u003e and (b) MH\u003c/p\u003e","description":"","filename":"15.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/a2dd826a9cc7f0b787830dbc.png"},{"id":89981533,"identity":"50784e81-6426-4533-b5c5-ad9152e60e27","added_by":"auto","created_at":"2025-08-27 06:25:49","extension":"png","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":491361,"visible":true,"origin":"","legend":"\u003cp\u003eSEM images showing various surface defects in SLM fabricated parts\u003c/p\u003e","description":"","filename":"16.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/b63849c83ff72a0963328d9d.png"},{"id":89981538,"identity":"59b01d50-6c2c-4e8f-a40b-77c7dcebb4b3","added_by":"auto","created_at":"2025-08-27 06:25:50","extension":"png","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":364206,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a\u003c/strong\u003e) Porosity mapping by Image J.\u003cstrong\u003e (b) \u003c/strong\u003ePorosity \u0026nbsp;values at different experimental runs\u003c/p\u003e","description":"","filename":"17.png","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/2fc1646020d858fdcc260dbd.png"},{"id":91465208,"identity":"765e5e53-0f62-4a7d-b925-d5a53e0a3532","added_by":"auto","created_at":"2025-09-16 18:31:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3376102,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7302973/v1/f296640b-bafb-4a8c-934b-ee4da401f4eb.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Prediction and Optimization of Surface quality and Microhardness using Machine learning in Selective laser melting of SS316L Biomedical alloy","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eThe fourth industrial revolution, or Industry 4.0, places a strong emphasis on automation, data-driven programming, and contemporary production methods. It highlights the challenge of achieving creative flexibility while building intricate structures that are valuable[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. A key technology in the present period, additive manufacturing (AM), often known as 3D printing, enables layer-by-layer product fabrication and digital manufacturing techniques. By linking, hardening, or depositing components, AM is a cutting-edge technique for converting CAD models into actual fabricated parts. Metal additive manufacturing (MAM) creates complex component designs by layering and melting metal powders using a high intensity beam [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. MAM methods can be divided into groups according on the energy source, such as laser beams or electrons. Selective Laser Melting (SLM), one of these laser-based MAMs, has achieved enormous traction in major industries including aerospace [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], medicinal [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], and automotive [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eMAM technology has made it possible to create complex components made of different types of materials. Steel is dominantly utilized material in SLM fabrication. Austenitic and low-carbon, AISI316L (SS316L) stainless steel (SS) prints exceptionally well in MAM [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. SS 316L is widely used in the medical field (implants, surgical equipment, etc.) due to its excellent biocompatibility; in the chemical, pharmaceutical, and marine industries because of its robust resistance to corrosion; and in the automotive and aerospace industries because of its suitable mechanical properties [\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eTucho and colleagues [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]examined the microstructure and defects of SS316L parts. It was observed complexity and indications of martensite or ferrite phases. The density of SLM fabricated 316L stainless steel increased, whereas microhardness and corrosion resistance decreased, according to Lin et al.[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] findings. Huang et al. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] discovered that energy density is the primary factor influencing the mechanical properties and density of SLMed components. They also discovered that an increase in hatch spacing (HS) led to the formation of voids, which reduces the density. According to research by Deng et al.[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], the relative density and Ra of SLM fabricated parts are significantly impacted by laser power (LP) and scanning speed (SS), whereas HS has a minor effect. According to Larimian et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], scanning technique, SS, and energy density all have an impact on the mechanical properties, microstructure, and densification of 316L alloy under study. Cherry et. al.[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] revealed that material hardness rises with VED up to 225 HV at 125 J/mm\u003csup\u003e3\u003c/sup\u003e, whereas further rise resulting in a decrease in hardness. Yusuf et. al.[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] discovered that the mean micro hardness of SLM samples in the x-y, x-z, and y-z planes are 262 HV, 237 HV, and 239 HV, correspondingly. The maximum micro hardness was found in the x-y plane, demonstrating anisotropy in SLM, which is characteristic of AM methods for metal elements. This is owing to the layer-wise construction method combined with the \"island\" scan strategy, leading to heterogeneous morphologies and asymmetric grain formations.\u003c/p\u003e\u003cp\u003ePagac et. al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] concluded that heat treatment and laser performance have no major influence on microhardness of the SLM fabricated components. SLM fabricated SS 316 L specimens is around 50 HBW harder, possibly owing to grain size, chemical improvements as well as internal tension. These parameters influence the component's hardness, according to the mean grain size in rolled steel 316L. Manikandan et. al.[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] discovered that micro hardness values grew along with rising VED, with a larger association reported for surface PPBD than surface PBD. Surface PPBD had the greatest microhardness value at 69 J/mm\u0026sup3;. Surface PBD had the highest at 308.8\u0026thinsp;\u0026plusmn;\u0026thinsp;8.7 HV and the lowest at 292.7\u0026thinsp;\u0026plusmn;\u0026thinsp;15.7 HV. Barrionuevo et. al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] discovered a direct correlation among microhardness and relative density, with higher density leading in increased micro hardness. The mean micro hardness found was approximately 237\u0026thinsp;\u0026plusmn;\u0026thinsp;9 HV. Nevertheless, outcomes varied, with the only evident connection being power and speed, with more power and lower speed leading in greater micro hardness.\u003c/p\u003e\u003cp\u003eA rough surface appearance caused by laser energy input, powder particle size, and material properties is the result of defect formation in SLMed parts, including lack of fusion, residual unmelted powder, open pores, material stacking, and balling effects. These defects may result in interlayer and multilayer defects [\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. LP and SS are key factors that influence the Ra of SLM-fabricated parts, which are influenced by track width, scan space, and powder layer thickness [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Surface quality may suffer from an excessive laser energy density. The R\u003csub\u003ea\u003c/sub\u003e of the component is affected by the wiper motion, gas flow, and scan direction [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. In LPBF process, R\u003csub\u003ea\u003c/sub\u003e value frequently surpasses 10 \u0026micro;m, while in mechanical processes like milling and grinding, it is between 1\u0026ndash;2 \u0026micro;m. Because this roughness serves as a site for crack formation, AM components have low fatigue limits [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Surface improvement parameters are frequently adjusted in the study conducted by Galy et al. [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] according to the laser energy input.\u003c/p\u003e\u003cp\u003eMachine learning (ML) enables machines to learn and perform operations autonomously. ML is categorized into supervised learning, unsupervised learning, and reinforcement learning [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. ML techniques enable effective data-driven strategies for maximizing the quality of SLM components by process parameter optimization, microstructure prediction and defect detection [\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. However, the efficiency of these techniques relies on adequacy of collected data through in-situ sensors, feature extraction and filtering of data collected [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Enhancing the efficiency and excellence of certain components is a vital research goal in the manufacturing process using the SLM technology. Moreover, the integration of Artificial Intelligence (AI) and ML algorithms may enhance the optimization of process signatures in real-time, resulting in accelerated and more effective production in L-PBF. An ensemble ML technique called Random Forest (RF) constructs several decision trees during training and aggregates their predictions to increase accuracy and decrease over-fitting. Gradient Boosting (GB), a progressive ML technique, improves classification and regression prediction accuracy by constructing trees to repair prior errors [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Extreme Gradient Boosting (XG boost) is a well-liked and successful use of the GB technique for regression predictive modeling. It is often utilized in ML modeling because of its great performance and fast computation speed [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. In order to anticipate and optimize process parameters, numerous studies have been carried out on the SLM technology and ML utilizing various materials [\u003cspan additionalcitationids=\"CR37 CR38 CR39 CR40 CR41\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e].\u003c/p\u003e\u003cp\u003ePrediction and optimization of process parameters during SLM fabrication of SS316L alloy utilizing tree-based boosting ML models has not been extensively studied till now. These boosting strategies use ensemble learning to construct a strong classifier from several weak classifiers for accurate prediction of R\u003csub\u003ea\u003c/sub\u003e and MH values. This study uses training and testing data sets to predict R\u003csub\u003ea\u003c/sub\u003e and MH values using Random Forest, Gradient Boosting, and XG Boosting techniques. The optimal ML model for maximizing the performance of SLM fabricated parts was determined by comparing experimental and predicted values. For detection and characterization of surface defects, SEM is used to capture images under different process settings. Image J software is also utilized for image processing and analysis of the SEM images for porosity mapping and quantification at different energy densities during SLM fabrication.\u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 SS 316L alloy\u003c/h2\u003e\u003cp\u003eFor the sample preparation, powdered austenitic stainless steel AISI 316L alloy with a particle size of 15\u0026ndash;60 \u0026micro;m was utilized. The chrome-nickel austenitic alloy is known as SS316L. Additionally, it contains molybdenum, which improves resistance to pitting from chloride ion solutions, boosts strength at high temperatures, and increases general corrosion resistance. Additionally, alloy 316L offers superior formability and fabricability. This alloy used in medical implants, architecture items, heat exchangers, fasteners, marine use, and in food equipment etc[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe chemical composition of the SS316L powder is shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eChemical composition of SS316L powder\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"12\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eElem.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFe\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCr\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNi\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMo\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMn\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSi\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eO\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e\u003cp\u003eCu\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c11\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c12\"\u003e\u003cp\u003eS\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWt %\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBal.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.005\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Experimental procedure\u003c/h2\u003e\u003cp\u003eThe SLM 280 is a selective laser melting machine utilized in this work for fabrication of components made of SS316L alloy. It has a beam diameter of 80 \u0026micro;m, a fiber laser type, and a power of 400W. This SLM machine was employed to fabricate cylindrical samples, as seen in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, with a diameter of 10 mm and a height of 5.0 mm. A CAD model of required shape and size has been prepared and exported to the SLM machine for fabrication using SS316L alloy powder (particle size: 30 \u0026micro;m). The LP, SS and HS are used as variable parameters while other parameters were kept as constant. All the experiments have been conducted using L\u003csub\u003e27\u003c/sub\u003e Taguchi design of experiment within the range of selected parameters. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e displays the range of process parameters that were chosen for performing SLM fabrication of SS316L alloy.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eRange of process parameters\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eS No\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eParameters\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e\u003cp\u003eRange\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLaser power (LP)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e225\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e275\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e325\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eScanning speed (SS)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e650\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e850\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHatch spacing (HS)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Measurement of surface roughness\u003c/h2\u003e\u003cp\u003eThe unevenness of a machined surface, which is defined by tiny peaks and valleys that are closely spaced, is referred to as the R\u003csub\u003ea\u003c/sub\u003e in machining. It plays a crucial role in forecasting the surface quality of components since it has a big impact on how well the parts function during fatigue loading. Durability and performance are adversely affected by surface asperities. As seen in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the R\u003csub\u003ea\u003c/sub\u003e of the manufactured samples was determined using a surface roughness tester [Make: Mitutoyo Japan; Make: SJ-201]. To increase measurement precision and decrease error, each observation was taken thrice, and the average value was calculated.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Measurement of microhardness\u003c/h2\u003e\u003cp\u003eMicrohardness (MH) is a material's hardness tested by forcing an indenter, such as a Vickers or Knoop indenter, into its surface under a stress ranging from 15 to 1000 gf; the indentations are often so minute that they must be determined with a microscope. The resin mold samples of required dimensions have been prepared for testing purposes. The Micro Wizhard Microhardness testing machine Ver.1.401 (Mitutoyo: Japan) is used for measurement of MH of SLM fabricated samples. The 980 gf indentation load has been applied for a dwell period of 15 seconds and using grid pattern as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Machine learning models\u003c/h2\u003e\u003cp\u003eIn the present work, three ensemble ML models have been utilized for prediction of R\u003csub\u003ea\u003c/sub\u003e and MH values and performance comparison at different process settings. Jupyter NoteBook and Python scripting were used for preprocessing and data modeling of three learning algorithms as shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The experimental data for R\u003csub\u003ea\u003c/sub\u003e and MH were collected and utilized to train three ML algorithms such as Random forest (RF), Gradient booting (GB) and XG boosting (XGB) to simulate the relationship between the input parameters and output response to reliably forecast R\u003csub\u003ea\u003c/sub\u003e and MH values for unknown input parameters like LP, SS and HS. MAE, MSE, RMSE and MPE are effectiveness measures for ML models. The details of RF, GB, and XG boost ML model parameters are shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDifferent libraries used in Python\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eS No\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eName of library\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePandas\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003esklearn.ensemble.GradientBoostingRegressor\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003esklearn.ensemble.AdaBoostRegressor\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003exgboost.XGBRegressor\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003esklearn.metrics\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumpy\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ematplotlib.pyplot\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDetails of RF, GBM and XG boost model parameters\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParameters/Methods\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGBM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eXG boost\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003en_estimators\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003emax_depth\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e---\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLearning rate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003erandom_state\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Evaluation of Machine learning models\u003c/h2\u003e\u003cp\u003eIn this study, statistical tests were performed to assess the prediction accuracy of the developed ML models. These measures include MSE, MAE and RMSE. These metrics are calculated using the equations below.\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:MSE=\\frac{1}{N}\\underset{i=1}{\\overset{N}{?}}{e}_{i}^{2}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:MAE=\\frac{1}{N}\\underset{i=1}{\\overset{N}{?}}\\left|{e}_{i}\\right|$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\:RMSE=\\sqrt{\\frac{1}{N}\\underset{i=1}{\\overset{N}{?}}{e}_{i}^{2}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn this case, N is the number of testing data samples, and e\u003csub\u003ei\u003c/sub\u003e is the error difference between the experimental and predicted R\u003csub\u003ea\u003c/sub\u003e and MH values for each data set. MSE, MAE, and RMSE are metrics used to monitor and evaluate the predicted models performance and accuracy.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results and Discussion","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.1 XRD Analysis\u003c/h2\u003e\u003cp\u003eX-ray diffraction (2D-XRD) was performed using a Bruker D8 ADVANCE. The samples were examined using a Cu K/α radiation (λ\u0026thinsp;=\u0026thinsp;1.54 \u0026Aring;) source at a voltage of 40 kV and current of 15 mA. The scan speed was 6 degrees/minute throughout a range of 5\u003csup\u003eo\u003c/sup\u003e to 100\u003csup\u003eo\u003c/sup\u003e. It also shows the existence of a single austenitic FCC phase, with no foreign phase peaks detected. The crystallographic planes {111}, {200}, {220}, and {311} were at 2θ angles of 43.60\u0026deg;, 50.759\u0026deg;, 74.595\u0026deg;, and 90.487\u0026deg;[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e], correspondingly as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Comparison of experimental and predicted surface roughness\u003c/h2\u003e\u003cp\u003eThe comparison of experimental and predicted values of R\u003csub\u003ea\u003c/sub\u003e for training and testing data sets is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. Based on the graph, Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows results of different ML Models and Error Metrics in terms of MAE, MSE and RMSE. It was found that values of MAE, MSE and RMSE for XG boosting were 0.73, 1.37 and 1.21 which are significantly lower values as compared to RF and GB methods. XG boosting showed 29.51% and 19.88% better accuracy in prediction of R\u003csub\u003ea\u003c/sub\u003e values as compared to RF and GB methods, respectively.\u003c/p\u003e\u003cp\u003eThe average error between experimental and predicted values of R\u003csub\u003ea\u003c/sub\u003e for RF and GB methods for the training and testing data sets were 2.45% and 2.16% respectively as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. In contrast, the respective values for XG boost model were 1.73%, respectively which is considerably lower value as compared to RF and GB model as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e. This is because of its better data handling capacity and efficient capturing of complex data sets [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. The minimum value of R\u003csub\u003ea\u003c/sub\u003e were found to be 9.41\u0026micro;m at optimum parameter settings i.e. 225W LP, 850 mm/s SS and 0.10 mm HS as shown by S/N ratio plot for R\u003csub\u003ea\u003c/sub\u003e in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e. A 29.64% reduction in R\u003csub\u003ea\u003c/sub\u003e value is achieved at optimum parameter settings as compared to worst setting.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison of ML Models and Error Metrics for R\u003csub\u003ea\u003c/sub\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMAE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMPE\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRandom forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.83\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.61\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eXG Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Comparison of experimental and predicted microhardness\u003c/h2\u003e\u003cp\u003eThe comparison of experimental and predicted values of MH for training and testing data sets is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e. Based on these graph, Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows comparison of different ML Models and Error Metrics in terms of MAE, MSE and RMSE. It was found that values of MAE, MSE and RMSE for XG boosting were 0.6687, 1.4478 and 1.2032 which are significantly lower values as compared to RF and GB methods. XG boosting showed 38.50% and 23.07% better accuracy in prediction of MH values as compared to RF and GBM methods, respectively.\u003c/p\u003e\u003cp\u003eThe average error between experimental and predicted values of MH for RF and GB methods for the training and testing data sets were 0.6465% and 0.3519% respectively as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e. In contrast, the respective values for XG boost model were 0.1217%, which is a considerably lower value as compared to RF and GB model as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e. This is because of its better data handling capacity and efficient capturing of complex data sets [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. The maximum value of MH was found to be 241.80 HV at optimum parameter settings i.e. 275W LP, 650 mm/s SS and 0.10 mm HS as shown by S/N ratio plot for MH in Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003e. A 14.73% improvement in microhardness is achieved at optimum parameter settings as compared to worst setting.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison of ML Models and Error Metrics for MH\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMAE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRMSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMPE\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRandom Forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.0642\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.8028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.3427\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.26\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.8296\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.4920\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.2802\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eXG Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.6687\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.4478\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.2032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Feature correlation heatmap\u003c/h2\u003e\u003cp\u003eIn ML technique, feature correlation heatmap is an effective visual aid for comprehending the connections among various features in a dataset. Each cell's color indicates the strength and direction of the connection, and it shows the correlation coefficients between every pair of features in a matrix. Figure\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e15\u003c/span\u003e(a) shows that LP has a strong linear relationship with R\u003csub\u003ea\u003c/sub\u003e, while SS had a negative correlation with R\u003csub\u003ea\u003c/sub\u003e, demonstrating an inverse relationship. Figure\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e15\u003c/span\u003e(b) shows that SS and HS had a strong negative correlation with MH, demonstrating an inverse relationship. While LP has no significant relationship with MH.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Analysis of variance\u003c/h2\u003e\u003cp\u003eThe analysis of variance (ANOVA) has been conducted to assess variation within and between groups for R\u003csub\u003ea\u003c/sub\u003e for smaller-the-better S/N ratios and for MH for larger-the-better, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. Based on the p-values (for 95% confidence interval), LP has the most dominant effect on R\u003csub\u003ea\u003c/sub\u003e as shown in Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. Lower LP, high SS and HS spacing is required to achieve minimum value of R\u003csub\u003ea\u003c/sub\u003e. At the optimum setting, R\u003csub\u003ea\u003c/sub\u003e is found to be 9.705 \u0026micro;m which is 29.60.5% lower than its value at worst parameter settings (13.832 \u0026micro;m). Moderate LP, lower SS and lower HS is required to achieve maximum value of MH. At the optimum parameter settings, MH is found to be 241.80 HV which is 14.73% higher than its value at worst parameter settings (207.37 HV). Medium LP, lower SS and lower HS is required to achieve minimum value of porosity. Based on the p-values (for 95% confidence interval), SS has the most dominant effect on MH value as shown in Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Ranking of process parameters\u003c/h2\u003e\u003cp\u003eThe ranking of each factor is based on delta statistics that compare the relative magnitude of the effects, as indicated in Tables\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e and \u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e, respectively. This includes the average of S/N ratio of each feature at each factor level. The delta value is obtained by subtracting the lowest average value of each factor from its greatest average value. The LP (delta value: 2.15) has the most significant effect on the R\u003csub\u003ea\u003c/sub\u003e followed by SS (delta value: 0.63) and HS (delta value: 0.12) [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. According to the S/N ratio values, SS (delta value: 0.45) has the largest impact on MH, subsequent to HS (delta value: 0.37). LP has the least impact on the value of MH of the SLM fabricated components (delta value: 0.05). The SS has the most dominant effect on MH values followed by HS and LP.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eANOVA table for R\u003csub\u003ea\u003c/sub\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSource\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSS\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMS\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLaser power\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e43.2920\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e21.6460\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e51.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScanning speed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.8714\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.4357\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.053\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHatch spacing\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.1154\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.0577\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.873\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.4366\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.4218\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e54.7153\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003eR-sq: 95.84, R-sq (adj): 94.77 (for 95% CI)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eStatistical ranking table for R\u003csub\u003ea\u003c/sub\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLevel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLaser\u003c/p\u003e\u003cp\u003epower\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eScanning\u003c/p\u003e\u003cp\u003espeed\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eHatch\u003c/p\u003e\u003cp\u003espacing\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-20.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-22.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-21.72\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-22.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-21.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-21.75\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-22.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-21.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-21.84\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDelta\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRank\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eANOVA table for MH\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSource\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSS\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMS\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLaser power\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e4.233\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.107\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScanning speed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e618.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e309.150\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e182.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHatch spacing\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e387.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e193.532\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e114.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e33.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.694\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1047.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003eR-sq: 96.77, R-sq (adj): 95.80 (for 95% CI)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eStatistical ranking table for MH\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLevel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLaser power\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eScanning speed\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eHatch spacing\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e46.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46.92\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e46.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46.73\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e46.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46.55\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDelta\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.37\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRank\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.7 Surface defects and porosity mapping\u003c/h2\u003e\u003cp\u003eVarious surface defects in SLM fabricated parts are find out using SEM (Make: Jeol, Model: JSM6610LV) at different process settings as depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e16\u003c/span\u003e. Keyhole porosity, binding porosity, lack of fusion, balling, cracking, and other surface imperfections can significantly affect the mechanical properties and performance of SLM fabricated parts. It is possible that trapped gases in the molten pool are the cause of gas porosity. When there is insufficient melting and bonding within a layer or between powder layers, there is no fusion. Inadequate scanning or insufficient energy intake may be responsible for occurrence of such types of defects during SLM fabrication. Keyhole porosity, which results in the formation of a gas cavity during melting, can occur when excessive laser energy is supplied during SLM fabrication of SS316 alloy.\u003c/p\u003e\u003cp\u003eThese SEM images were further treated by Image J software for digital image processing and porosity mapping as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e17\u003c/span\u003e(a). This is open source software operating on Java platform for processing and analyzing scientific images [imagej.net]. The maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different volume energy densities (VEDs) after image processing by image J software as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e17\u003c/span\u003e(b).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Conclusions","content":"\u003cp\u003eIn this work, three tree based ML models such as Random forest, Gradient boosting and XG boosting are used for prediction and optimization of R\u003csub\u003ea\u003c/sub\u003e and MH values of SLM fabricated parts made of SS316L alloy. Following conclusions have been obtained from the present research:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThese ensemble learning ML models can provide better accuracy and reliability for predictive modeling for R\u003csub\u003ea\u003c/sub\u003e and MH under different process settings.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe MAE, MSE, and RMSE statistical values obtained from testing data are found to be 0.73, 1.37, and 1.21, for R\u003csub\u003ea\u003c/sub\u003e values.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe MAE, MSE, and RMSE statistical values obtained from testing data are found to be 0.6687, 1.4478, and 1.2032, for MH values.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe testing process revealed that the closest result to the experimental values was delivered by the XG booting model.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe average error of XG boost model is 0.1217% and 1.73%, respectively which is significantly lower as compared to RF and GB methods.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eTherefore, XG boosting showed better accuracy in prediction of R\u003csub\u003ea\u003c/sub\u003e and MH values as compared to Random forest and Gradient boosting methods.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eXG boosting showed 29.51% and 19.88% better accuracy in prediction of R\u003csub\u003ea\u003c/sub\u003e values as compared to RF and GB methods.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eXG boosting showed 38.50% and 23.07% better accuracy in prediction of MH values as compared to RF and GBM methods.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eHence, XG boosting can be adopted for the prediction of R\u003csub\u003ea\u003c/sub\u003e and MH values of SLM fabricated samples made of SS316L powder.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eA 29.64% reduction in R\u003csub\u003ea\u003c/sub\u003e value is obtained at optimum process settings as compared to worst setting. This will improve the performance and durability of fabricated parts.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eA 14.73% improvement in MH value is obtained at optimum process settings as compared to worst setting.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eThe maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different energy densities by image processing by image J software.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMost of the pores were found spherical shaped due to gas atomization of SS316L powder.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eUsing ML models in metal additive manufacturing will help enhance process control, decrease trial and error, and increase the quality and longevity of the final SLM fabricated parts.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical Approval declaration is not applicable. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding was received for this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA.S.: conceptualization, formal analysis, methodology, investigation, resources and writing\u0026mdash;original draft; T.U.S: formal analysis, investigation, writing\u0026mdash;review and editing, supervision and validation; M.K.S.: investigation, writing\u0026mdash;review and editing, supervision and validation; A.N.S.: supervision, methodology, writing\u0026mdash;review and editing, and validation; T.B.: methodology, formal analysis, investigation, resources and writing\u0026mdash;original draft\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eButt, J.: Exploring the Interrelationship between Additive Manufacturing and Industry 4.0. Designs. \u003cb\u003e4\u003c/b\u003e, 13 (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/designs4020013\u003c/span\u003e\u003cspan address=\"10.3390/designs4020013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDilberoglu, U.M., Gharehpapagh, B., Yaman, U., Dolen, M.: The Role of Additive Manufacturing in the Era of Industry 4.0. Procedia Manuf. \u003cb\u003e11\u003c/b\u003e, 545\u0026ndash;554 (2017). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.promfg.2017.07.148\u003c/span\u003e\u003cspan address=\"10.1016/j.promfg.2017.07.148\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAversa, A., Saboori, A., Marchese, G., et al.: Recent Progress in Beam-Based Metal Additive Manufacturing from a Materials Perspective: A Review of Patents. J. Materi Eng. Perform. \u003cb\u003e30\u003c/b\u003e, 8689\u0026ndash;8699 (2021). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11665-021-06273-3\u003c/span\u003e\u003cspan address=\"10.1007/s11665-021-06273-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStolt, R., Elgh, F.: Introducing design for selective laser melting in aerospace industry☆. J. Comput. Des. Eng. \u003cb\u003e7\u003c/b\u003e, 489\u0026ndash;497 (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/jcde/qwaa042\u003c/span\u003e\u003cspan address=\"10.1093/jcde/qwaa042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTalib Mohammed, M.: Mechanical Properties of SLM-Titanium Materials for Biomedical Applications: A Review. Materials Today: Proceedings 5:17906\u0026ndash;17913. (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.matpr.2018.06.119\u003c/span\u003e\u003cspan address=\"10.1016/j.matpr.2018.06.119\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang, J., Li, B., Liu, J., et al.: Application of Additive Manufacturing in the Automobile Industry: A Mini Review. Processes. \u003cb\u003e12\u003c/b\u003e, 1101 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/pr12061101\u003c/span\u003e\u003cspan address=\"10.3390/pr12061101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu, M., Guo, H., Wang, Y., et al.: Mechanical properties and microstructural characteristics of 316L stainless steel fabricated by laser powder bed fusion and binder jetting. J. Mater. Res. Technol. \u003cb\u003e24\u003c/b\u003e, 4427\u0026ndash;4439 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmrt.2023.04.069\u003c/span\u003e\u003cspan address=\"10.1016/j.jmrt.2023.04.069\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, N., Huang, S., Zhang, G., et al.: Progress in additive manufacturing on new materials: A review. J. Mater. Sci. Technol. \u003cb\u003e35\u003c/b\u003e, 242\u0026ndash;269 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmst.2018.09.002\u003c/span\u003e\u003cspan address=\"10.1016/j.jmst.2018.09.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeng, T., Chen, C.: Influence of energy density on energy demand and porosity of 316L stainless steel fabricated by selective laser melting. Int. J. Precis Eng. Manuf-Green Tech. \u003cb\u003e5\u003c/b\u003e, 55\u0026ndash;62 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40684-018-0006-9\u003c/span\u003e\u003cspan address=\"10.1007/s40684-018-0006-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYap, C.Y., Chua, C.K., Dong, Z.L., et al.: Review of selective laser melting: Materials and applications. Appl. Phys. Reviews. \u003cb\u003e2\u003c/b\u003e, 041101 (2015). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1063/1.4935926\u003c/span\u003e\u003cspan address=\"10.1063/1.4935926\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTucho, W.M., Lysne, V.H., Austb\u0026oslash;, H., et al.: Investigation of effects of process parameters on microstructure and hardness of SLM manufactured SS316L. J. Alloys Compd. \u003cb\u003e740\u003c/b\u003e, 910\u0026ndash;925 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jallcom.2018.01.098\u003c/span\u003e\u003cspan address=\"10.1016/j.jallcom.2018.01.098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLin, K., Gu, D., Xi, L., et al.: Selective laser melting processing of 316L stainless steel: effect of microstructural differences along building direction on corrosion behavior. Int. J. Adv. Manuf. Technol. \u003cb\u003e104\u003c/b\u003e, 2669\u0026ndash;2679 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-019-04136-9\u003c/span\u003e\u003cspan address=\"10.1007/s00170-019-04136-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, M., Zhang, Z., Chen, P.: Effect of selective laser melting process parameters on microstructure and mechanical properties of 316L stainless steel helical micro-diameter spring. Int. J. Adv. Manuf. Technol. \u003cb\u003e104\u003c/b\u003e, 2117\u0026ndash;2131 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-019-03928-3\u003c/span\u003e\u003cspan address=\"10.1007/s00170-019-03928-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng, Y., Mao, Z., Yang, N., et al.: Collaborative Optimization of Density and Surface Roughness of 316L Stainless Steel in Selective Laser Melting. Materials. \u003cb\u003e13\u003c/b\u003e, 1601 (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ma13071601\u003c/span\u003e\u003cspan address=\"10.3390/ma13071601\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLarimian, T., Kannan, M., Grzesiak, D., et al.: Effect of energy density and scanning strategy on densification, microstructure and mechanical properties of 316L stainless steel processed via selective laser melting. Mater. Sci. Engineering: A. \u003cb\u003e770\u003c/b\u003e, 138455 (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.msea.2019.138455\u003c/span\u003e\u003cspan address=\"10.1016/j.msea.2019.138455\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCherry, J.A., Davies, H.M., Mehmood, S., et al.: Investigation into the effect of process parameters on microstructural and physical properties of 316L stainless steel parts by selective laser melting. Int. J. Adv. Manuf. Technol. \u003cb\u003e76\u003c/b\u003e, 869\u0026ndash;879 (2015). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-014-6297-2\u003c/span\u003e\u003cspan address=\"10.1007/s00170-014-6297-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYusuf, S., Chen, Y., Boardman, R., et al.: Investigation on Porosity and Microhardness of 316L Stainless Steel Fabricated by Selective Laser Melting. Metals. \u003cb\u003e7\u003c/b\u003e, 64 (2017). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/met7020064\u003c/span\u003e\u003cspan address=\"10.3390/met7020064\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePag\u0026aacute;č, M., Hajnyš, J., Petrů, J., Zl\u0026aacute;mal, T.: Comparison of Hardness of Surface 316L Stainless Steel Made by Additive Technology and Cold Rolling. MSF. \u003cb\u003e919\u003c/b\u003e, 84\u0026ndash;91 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4028/www.scientific.net/MSF.919.84\u003c/span\u003e\u003cspan address=\"10.4028/www.scientific.net/MSF.919.84\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManikandan, P., Venkatesan, K.: Role of volumetric energy density on surface quality and mechanical properties of selective laser melted 310 stainless steel. Results Eng. \u003cb\u003e25\u003c/b\u003e, 104479 (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rineng.2025.104479\u003c/span\u003e\u003cspan address=\"10.1016/j.rineng.2025.104479\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarrionuevo, G.O., Ramos-Grez, J.A., S\u0026aacute;nchez-S\u0026aacute;nchez, X., et al.: Influence of the Processing Parameters on the Microstructure and Mechanical Properties of 316L Stainless Steel Fabricated by Laser Powder Bed Fusion. J. Manuf. Mater. Process. \u003cb\u003e8\u003c/b\u003e, 35 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/jmmp8010035\u003c/span\u003e\u003cspan address=\"10.3390/jmmp8010035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou, X., Wang, D., Liu, X., et al.: 3D-imaging of selective laser melting defects in a Co\u0026ndash;Cr\u0026ndash;Mo alloy by synchrotron radiation micro-CT. Acta Mater. 1\u0026ndash;16 (2015)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu, W.H., Sing, S.L., Chua, C.K., et al.: Particle-reinforced metal matrix nanocomposites fabricated by selective laser melting: A state of the art review. Prog. Mater. Sci. \u003cb\u003e104\u003c/b\u003e, 330\u0026ndash;379 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.pmatsci.2019.04.006\u003c/span\u003e\u003cspan address=\"10.1016/j.pmatsci.2019.04.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSharma, H., Singla, J., Singh, V., et al.: Influence of post heat treatment on metallurgical, mechanical, and corrosion analysis of wire arc additive manufactured inconel 625. J. Mater. Res. Technol. \u003cb\u003e27\u003c/b\u003e, 5910\u0026ndash;5923 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmrt.2023.11.074\u003c/span\u003e\u003cspan address=\"10.1016/j.jmrt.2023.11.074\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, D., Liu, Y., Yang, Y., Xiao, D.: Theoretical and experimental study on surface roughness of 316L stainless steel metal parts obtained through selective laser melting. RPJ. \u003cb\u003e22\u003c/b\u003e, 706\u0026ndash;716 (2016). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/RPJ-06-2015-0078\u003c/span\u003e\u003cspan address=\"10.1108/RPJ-06-2015-0078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, B.-Q., Li, Z., Bai, P., et al.: Research on Surface Roughness of AlSi10Mg Parts Fabricated by Laser Powder Bed Fusion. Metals. \u003cb\u003e8\u003c/b\u003e, 524 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/met8070524\u003c/span\u003e\u003cspan address=\"10.3390/met8070524\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbd-Elaziem, W., Elkatatny, S., Sebaey, T.A., et al.: Machine learning for advancing laser powder bed fusion of stainless steel. J. Mater. Res. Technol. \u003cb\u003e30\u003c/b\u003e, 4986\u0026ndash;5016 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmrt.2024.04.130\u003c/span\u003e\u003cspan address=\"10.1016/j.jmrt.2024.04.130\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGaly, C., Le Guen, E., Lacoste, E., Arvieu, C.: Main defects observed in aluminum alloy parts produced by SLM: From causes to consequences. Additive Manuf. \u003cb\u003e22\u003c/b\u003e, 165\u0026ndash;175 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.addma.2018.05.005\u003c/span\u003e\u003cspan address=\"10.1016/j.addma.2018.05.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMahmoud, D., Magolon, M., Boer, J., et al.: Applications of machine learning in process monitoring and controls of L-PBF additive manufacturing: A review. Appl. Sci. \u003cb\u003e11\u003c/b\u003e, 11910 (2021)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEshkabilov, S., Ara, I., Azarmi, F.: A comprehensive investigation on application of machine learning for optimization of process parameters of laser powder bed fusion-processed 316L stainless steel. Int. J. Adv. Manuf. Technol. \u003cb\u003e123\u003c/b\u003e, 2733\u0026ndash;2756 (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-022-10331-y\u003c/span\u003e\u003cspan address=\"10.1007/s00170-022-10331-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOkaro, I.A., Jayasinghe, S., Sutcliffe, C., et al.: Automatic fault detection for laser powder-bed fusion using semi-supervised machine learning. Additive Manuf. \u003cb\u003e27\u003c/b\u003e, 42\u0026ndash;53 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.addma.2019.01.006\u003c/span\u003e\u003cspan address=\"10.1016/j.addma.2019.01.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarrionuevo, G.O., Ramos-Grez, J.A., Walczak, M., Betancourt, C.A.: Comparative evaluation of supervised machine learning algorithms in the prediction of the relative density of 316L stainless steel fabricated by selective laser melting. Int. J. Adv. Manuf. Technol. \u003cb\u003e113\u003c/b\u003e, 419\u0026ndash;433 (2021). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-021-06596-4\u003c/span\u003e\u003cspan address=\"10.1007/s00170-021-06596-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCaggiano, A., Zhang, J., Alfieri, V., et al.: Machine learning-based image processing for on-line defect recognition in additive manufacturing. CIRP Ann. \u003cb\u003e68\u003c/b\u003e, 451\u0026ndash;454 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cirp.2019.03.021\u003c/span\u003e\u003cspan address=\"10.1016/j.cirp.2019.03.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDelli, U., Chang, S.: Automated Process Monitoring in 3D Printing Using Supervised Machine Learning. Procedia Manuf. \u003cb\u003e26\u003c/b\u003e, 865\u0026ndash;870 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.promfg.2018.07.111\u003c/span\u003e\u003cspan address=\"10.1016/j.promfg.2018.07.111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBiau, G., Cadre, B.: Optimization by Gradient Boosting. In: Daouia, A., Ruiz-Gazen, A. (eds.) Advances in Contemporary Statistics and Econometrics: Festschrift in Honor of Christine Thomas-Agnan, pp. 23\u0026ndash;44. Springer International Publishing, Cham (2021)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee, S., Park, J., Kim, N., et al.: Extreme gradient boosting-inspired process optimization algorithm for manufacturing engineering applications. Mater. Design. \u003cb\u003e226\u003c/b\u003e, 111625 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.matdes.2023.111625\u003c/span\u003e\u003cspan address=\"10.1016/j.matdes.2023.111625\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAqilah, D.N., Sayuti, A.K.M., Farazila, Y., et al.: Effects of Process Parameters on the Surface Roughness of Stainless Steel 316L Parts Produced by Selective Laser Melting. J. Test. Eval. \u003cb\u003e46\u003c/b\u003e, 1673\u0026ndash;1683 (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1520/JTE20170140\u003c/span\u003e\u003cspan address=\"10.1520/JTE20170140\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, J., Jeong, S.G., Kim, E.S., et al.: Material-agnostic machine learning approach enables high relative density in powder bed fusion products. Nat. Commun. \u003cb\u003e14\u003c/b\u003e, 6557 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-023-42319-x\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-42319-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTan, X., Chen, D., Xiao, H., et al.: Prediction of phase and tensile properties of selective laser melting manufactured high entropy alloys by machine learning. Mater. Today Commun. \u003cb\u003e41\u003c/b\u003e, 110209 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.mtcomm.2024.110209\u003c/span\u003e\u003cspan address=\"10.1016/j.mtcomm.2024.110209\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eToprak, C.B., Dogruer, C.U.: Optimal process parameter determination in selective laser melting via machine learning-guided sequential quadratic programing. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 239:807\u0026ndash;820. (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/09544062241288948\u003c/span\u003e\u003cspan address=\"10.1177/09544062241288948\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePashmforoush, F., Seyedzavvar, M.: A transfer learning-based machine learning approach to predict mechanical properties of different material types fabricated by selective laser melting process. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering 09544089231215683. (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/09544089231215683\u003c/span\u003e\u003cspan address=\"10.1177/09544089231215683\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYehia, H.M., Hamada, A., Sebaey, T.A., Abd-Elaziem, W.: Selective Laser Sintering of Polymers: Process Parameters, Machine Learning Approaches, and Future Directions. J. Manuf. Mater. Process. \u003cb\u003e8\u003c/b\u003e, 197 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/jmmp8050197\u003c/span\u003e\u003cspan address=\"10.3390/jmmp8050197\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChang, L.-K., Jiang, K., Chung, C., et al.: Integrating machine learning and multi-objective optimization to investigate the magnetic and mechanical properties of FeSiCr soft magnetic composite processed by selective laser melting. Int. J. Adv. Manuf. Technol. \u003cb\u003e132\u003c/b\u003e, 3637\u0026ndash;3653 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00170-024-13589-6\u003c/span\u003e\u003cspan address=\"10.1007/s00170-024-13589-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD\u0026rsquo;Andrea, D.: Additive Manufacturing of AISI 316L Stainless Steel: A Review. Metals. \u003cb\u003e13\u003c/b\u003e, 1370 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/met13081370\u003c/span\u003e\u003cspan address=\"10.3390/met13081370\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePuichaud, A.-H., Flament, C., Chniouel, A., et al.: Microstructure and mechanical properties relationship of additively manufactured 316L stainless steel by selective laser melting. EPJ Nuclear Sci. Technol. \u003cb\u003e5\u003c/b\u003e, 23 (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1051/epjn/2019051\u003c/span\u003e\u003cspan address=\"10.1051/epjn/2019051\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGadagi, A., Sivaprakash, B., Adake, C., et al.: Epoxy composite reinforced with jute/basalt hybrid \u0026ndash; Characterisation and performance evaluation using machine learning techniques. Compos. Part. C: Open. Access. \u003cb\u003e14\u003c/b\u003e, 100453 (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jcomc.2024.100453\u003c/span\u003e\u003cspan address=\"10.1016/j.jcomc.2024.100453\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\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":"Selective laser melting, SS316L alloy, Surface roughness, Microhardness, Machine learning","lastPublishedDoi":"10.21203/rs.3.rs-7302973/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7302973/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLaser additive manufacturing based selective laser melting (SLM) technique have attracted a lot of attention due to the rising need of high performance materials in aerospace, automotive, and biomedical applications. However, because of the intricate relationships between several parameters including laser power, scanning speed, hatch spacing, and layer thickness, optimization of the process parameters for SLM is a tedious task. Machine learning (ML) technique can handle a variety of data sets and can accurately predict complicated and non-linear relationships in SLM. In this paper, three tree based ML models such as Random forest, Gradient Boosting, and XG boost Regressor are used for prediction of surface roughness (R\u003csub\u003ea\u003c/sub\u003e) and microhardness (MH) of SLM fabricated parts for improved part quality and longevity. The efficacy of the ML models is evaluated in terms of prediction accuracy and computational efficiency after training and testing to predict optimal process parameters for minimum R\u003csub\u003ea\u003c/sub\u003e and maximum MH, respectively. The average error of XG boost model for prediction of R\u003csub\u003ea\u003c/sub\u003e and MH is 0.1217% and 1.73%, respectively which is significantly lower as compared to Random forest and Gradient boosting methods. Therefore, XG boosting showed better accuracy in prediction of R\u003csub\u003ea\u003c/sub\u003e and MH values as compared to Random forest and Gradient boosting methods. This is because of its better data handling capacity and efficient capturing of complex data sets. A 29.64% decrease in R\u003csub\u003ea\u003c/sub\u003e and 14.73% increase in MH values are achieved at optimized settings for performance improvement of SLM fabricated parts. The maximum and minimum porosity in SLM fabricated parts is found to be 0.987% and 0.249% at different energy densities after image processing by Image J software. This work will be useful in implementation of ML technique in SLM fabrication for better process control, reduction in trial-and-error, and to improve the functionality and reliability of finished parts.\u003c/p\u003e","manuscriptTitle":"Prediction and Optimization of Surface quality and Microhardness using Machine learning in Selective laser melting of SS316L Biomedical alloy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-27 06:17:43","doi":"10.21203/rs.3.rs-7302973/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":"c2621f4a-b25b-424c-b900-1ee18e196a35","owner":[],"postedDate":"August 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-09-16T18:23:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-27 06:17:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7302973","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7302973","identity":"rs-7302973","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}