108 Advances in Production Engineering & Management ISSN 1854-6250 Volume 19 | Number 1 | March 2024 | pp 108–116 Journal home: apem-journal.org https://doi.org/10.14743/apem2024.1.496 Original scientific paper FDM process parameter selection by hybrid MCDM approach for flexural and compression strength maximization Begic-Hajdarevic, D. a , Klančnik, S. b , Muhamedagić, K. a , Čekić, A. a , Cohodar Husic, M. a , Ficko, M. b,* , Gusel, L. b a University of Sarajevo, Faculty of Mechanical Engineering, Sarajevo, Bosnia and Herzegovina b University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia A B S T R A C T A R T I C L E I N F O Fused deposition modelling (FDM) is one of the mostly used additive technol- ogies, due to its ability to produce complex parts with good mechanical prop- erties. The selection of FDM process parameters is crucial to achieve good mechanical properties of the manufactured parts. Therefore, in this paper, a hybrid multi-criteria decision-making (MCDM) approach based on Preference Selection Index (PSI) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is proposed for the selection of optimal process pa- rameters in FDM printing of polylactic acid (PLA) parts. Printing temperature, layer thickness and raster angle were considered as input process parame- ters. In order to prove the effectiveness of the proposed hybrid PSI – TOPSIS method, the obtained results were compared with the results obtained with different MCDM methods. The obtained best option of process parameters was confirmed by other MCDM methods. The optimal combination of process parameters to achieve the maximal flexural strength, maximal flexural modu- lus and maximal compressive strength is selected using the hybrid PSI-TOPSIS method. The results show that the hybrid PSI-TOPSIS approach could be used for optimisation process parameters for any machining process. Keywords: Fused deposition modelling (FDM); Multi-criteria decision-making (MCDM); Hybrid PSI-TOPSIS method; Process parameters; Mechanical properties; Optimization *Corresponding author: mirko.ficko@um.si (Ficko, M.) Article history: Received 6 November 2023 Revised 6 March 2024 Accepted 13 March 2024 Content from this work may be used under the terms of the Creative Commons Attribution 4.0 International Licence (CC BY 4.0). Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. 1. Introduction Additive manufacturing (AM) represents a way of production that is based on making products by adding materials “layer by layer” This method of production was initially used only for the rapid prototyping (RP), but today additive technologies are also used for the production of high- ly functional products in small quantities. This principle of making parts remains the same re- gardless of the degree of geometric complexity of the part, which is the main advantage of this technology. According to the physics of the process and material type a large number of different additive manufacturing processes have been developed. Some of the most commonly used procedures are: fused deposition modelling (FDM), stereolithography (SLA), Ink jet modelling, selective la- ser sintering (SLS), etc. The FDM process is used for prototyping and production of fully func- tional parts for engineering applications. Some of most commonly used material for FDM pro- cess are: Polylactic-Acid (PLA), Polyethylene-Terephthalate (PET), Acrylonitrile Butadiene- FDM process parameter selection by hybrid MCDM approach for flexural and compression strength maximization Advances in Production Engineering & Management 19(1) 2024 109 Styrene (ABS), propylene (PP), Polyamide (PA), and Thermoplastic-Polyurethane (TPU) [1,2]. Quality and mechanical properties of FDM produced parts are key factors for their use in indus- trial applications. In order to achieve the appropriate quality and mechanical properties of parts made in this way, it is necessary to carefully design the FDM process in terms of the correct se- lection of input parameters. Due to the large number of input and output parameters, it is often necessary to solve complex optimization problems. In order to solve such problems and avoid the need to perform a large number of experiments, a systematic approach to the experiment plan and the application of various methods of multi-criteria optimization are used. Some of the most commonly used methods for experiment design, modelling and optimization of process parameters of the FDM technology are: Taguchi Method [3], Grey Relational Analysis (GRA) [4], Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) [5], Response Surface Methodology (RSM) [6], Genetic Algorithm (GA) [7], Artificial Neural Network (ANN) [6], and Adaptive Neuro Fuzzy Interface System (ANFIS) [8]. Many researchers have analysed the possi- bility of applying these methods to optimize the various output parameters of the FDM process. In study [9], the authors optimized three input parameters, infill density, printing speed and printing temperature, to achieve the maximum tensile strength of samples made of PLA materi- al. For optimisation these process parameters were used hybrid optimization techniques, genet- ic algorithm-artificial neural network, genetic algorithm-response surface methodology and ge- netic algorithm-adaptive neuro fuzzy interface system. It is shown that such hybrid models could be used for optimisation any other process parameters for any industrial application prob- lems. Rajamani et al. [5] were used hybrid approach through RSM-TOPSIS method. This ap- proach was used for improving surface quality of micro sized near-net-shaped components for end use applications using FDM additive manufacturing techniques. Production of test speci- mens was carried out according to the previously defined Box-Behnken experimental design. For input parameters were selected: layer thickness, part orientation, raster width and raster angle. The optimal FDM parameters for improved surface quality attributes were determined using TOPSIS method. This method proved to be a useful tool for finding optimal FDM process parameters for fabricating the components of a flapping wing micro mechanism. Also, the TOP- SIS method proved to be useful in the selection of optimal process parameters in two-point in- cremental forming process [10], as well as for optimization of cutting parameters in turning pro- cess [11], and thanks to the proposed Fuzzy-TOPSIS approach [12], managers of manufacturing companies can access and monitor the maintenance sustainability level integrated with the in- dustry 4.0 technologies. The RSM method is a combination of statistical and mathematical meth- ods that is very useful for modelling and optimizing engineering scientific problems, which gives very low standard errors to experimental verification. Srinivasan et al. [13] used the RSM meth- od based on central composite design to predict of tensile strength in FDM printed ABS parts. In paper [6] was successfully applied RSM and ANN to investigate the effect of the layer thickness, printing speed, raster angle and wall thickness on the tensile strength of test specimens printed with a short carbon fibre reinforced polyamide composite. Taguchi-Grey relational analysis was used in the study [14] to optimize input parameters and improve selected output mechanical characteristics. This study is designed to capture the said gap in the literature with focus on cell geometry, nozzle diameter and strain rate by using the Taguchi design of experimentation and Grey Relational Analysis. It is shown that the GRA meth- od significantly simplifies complex optimization problems in FDM process parameters optimiza- tion. Taguchi method is very useful in the experimental plan phase, and it can be used separately [15] or with other methods for multi-criteria optimization [14] [16]. Chohan et al. [16] were using Taguchi-TOPSIS based optimization of FDM process parameters for manufacturing ABS plastics parts. The results were shown that using the TOPSIS method, optimal parameters can be determined in order to improve the surface-quality of FDM parts which can be utilized for end- use products and for rapid tooling applications. In addition to the mentioned methods, there is also the Preference Selection Index (PSI) method that can be used to solve multi-criteria optimization problems. The possibility of apply- ing the PSI method for the selection of optimal FDM process parameters was investigated [17]. It is found that the PSI method is very simple to understand and easy to implement. The advantage Begic-Hajdarevic, Klančnik, Muhamedagić, Čekić, Cohodar Husic, Ficko, Gusel 110 Advances in Production Engineering & Management 19(1) 2024 of the PSI method is that there is no need to calculate the relative weight of outputs. However, some authors [18] have observed that this method is not useful when several alternatives have criteria values that are very close to those are preferred. A hybrid TOPSIS-PSI method for selec- tion material in marine applications was presented in study [19], the entropy method has been used to determine the weights of the selected criteria. In this paper, a hybrid method that combines of the PSI and TOPSIS method is proposed. The proposed hybrid method considered the advantage of the PSI method that does not require the calculation of the weight factor of criteria and the advantage of the TOPSIS method that is more efficient in dealing with the criteria and the number of available alternatives. To test the pro- posed method, the case of selecting optimal process parameters to improve the mechanical properties of FDM printed PLA parts was considered. 2. Materials and methods 2.1 Experimental details A 3D printer Ultimaker S5 was used to produce test samples from PLA material, that is one of the mostly used FDM material. The samples for flexural and the compressive tests were designed and tested according to ISO 178 and ISO 604 standards, respectively. The constant process pa- rameters for printing test samples are shown in Table 1. In this paper, three input process pa- rameters, namely printing temperature, layer thickness and raster angle were investigated in order to study their influence on the mechanical properties of the test samples using the Taguchi design of experiments. These parameters are varied at three different levels as shown in Table 2. Table 1 The constant process parameters and their values Parameter Unit Value Nozzle diameter mm 0.4 Infill density % 100 Build plate temperature °C 110 Build direction Flat x-x direction Printing speed mm/s 60 Table 2 The input parameters and their levels Parameter Symbol Unit Level 1 Level 2 Level 3 Printing temperature T °C 180 200 220 Layer thickness L mm 0.1 0.2 0.3 Raster angle A ° 0 45 90 Three samples were tested for each set of input parameters. The experimentally studied out- put parameters were flexural strength (FS), flexural modulus (FM) and compressive strength (CS). The previous research [2, 6-9] were focused on analysing the influence of process parame- ters on the tensile strength of FDM printed parts. The average value of the output parameters is reported in Table 3. Flexural and compressive tests were conducted on the 10 kN Shimadzu AGS-X universal machine. Table 3 Experimental data Exp. No. Input parameters Output parameters L (mm) T (°C) A (°) Flexural strength (MPa) Flexural modulus (MPa) Compressive strength (MPa) 1 0.1 180 0 38.55 2505.65 45.16 2 0.1 200 45 81.56 3014.92 46.78 3 0.1 220 90 92.24 2952.91 48.14 4 0.2 180 45 36.53 1794.45 39.45 5 0.2 200 90 79.77 2598.18 45.34 6 0.2 220 0 72.24 2650.26 44.37 7 0.3 180 90 29.65 2054.71 39.81 8 0.3 200 0 52.10 2221.84 41.63 9 0.3 220 45 72.38 2502.59 41.30 FDM process parameter selection by hybrid MCDM approach for flexural and compression strength maximization Advances in Production Engineering & Management 19(1) 2024 111 2.2 Hybrid PSI-TOPSIS method Process parameters selection for any machining process is a MCDM problem that considers dif- ferent competing criteria for selecting appropriate process parameters. The proposed hybrid PSI – TOPSIS method consists of the following steps. Step 1: Determine a set of experimental trials (alternatives): 𝐸𝐸 = [ 𝐸𝐸 1 , 𝐸𝐸 2 , … , 𝐸𝐸 𝑚𝑚 ] (1) where m is the number of experimental trials. Step 2: Determine a set of criteria (output parameters): 𝐶𝐶 = [ 𝐶𝐶 1 , 𝐶𝐶 2 , … , 𝐶𝐶 𝑛𝑛 ] (2) where n is the number of criteria. Step 3: Creating a decision matrix: 𝐷𝐷 = � 𝐷𝐷 𝑖𝑖𝑖𝑖 | 𝑖𝑖 = 1, 2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 � (3) and 𝐷𝐷 𝑖𝑖𝑖𝑖 is the value of the j-th criterion for the i-th experimental trial. Step 4: Calculation of the normalized matrix: a) if the larger is better (LB): 𝑁𝑁 𝑖𝑖𝑖𝑖 = 𝐷𝐷 𝑖𝑖𝑖𝑖 𝐷𝐷 𝑖𝑖 ma x , 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (4) b) if the smaller is better (SB): 𝑁𝑁 𝑖𝑖𝑖𝑖 = 𝐷𝐷 𝑖𝑖 m in 𝐷𝐷 𝑖𝑖𝑖𝑖 , 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (5) Step 5: Calculating the mean value of the normalized matrix: 𝑁𝑁 𝑖𝑖 = 1 𝑚𝑚 � 𝑁𝑁 𝑖𝑖𝑖𝑖 𝑚𝑚 𝑖𝑖 = 1 , 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (6) Step 6: Calculating the value of the preference variation: 𝜑𝜑 𝑖𝑖 = � � 𝑁𝑁 𝑖𝑖𝑖𝑖 − 𝑁𝑁 𝑖𝑖 � 2 𝑚𝑚 𝑖𝑖 = 1 , 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (7) Step 7: Calculating the deviation in preference value: ∆ 𝑖𝑖 = � 1 − � � 𝑁𝑁 𝑖𝑖𝑖𝑖 − 𝑁𝑁 𝑖𝑖 � 2 𝑚𝑚 𝑖𝑖 = 1 � , 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (8) Step 8: Determine the overall preference value (weight factors for each criteria): 𝑝𝑝 𝑖𝑖 = ∆ 𝑖𝑖 ∑ ∆ 𝑖𝑖 𝑛𝑛 𝑖𝑖 = 1 , 𝑗𝑗 = 1,2, … , 𝑛𝑛 and � 𝑝𝑝 𝑖𝑖 𝑛𝑛 𝑖𝑖 = 1 = 1 (9) Step 9: Creating a weighted normalized decision matrix: 𝑤𝑤 𝑖𝑖𝑖𝑖 = 𝑝𝑝 𝑖𝑖 × 𝑁𝑁 𝑖𝑖𝑖𝑖 for 𝑖𝑖 = 1,2, … , 𝑚𝑚 ; 𝑗𝑗 = 1,2, … , 𝑛𝑛 (10) Step 10: Determine the positive (PS) and negative ideal solution (NS): 𝑃𝑃 𝑃𝑃 = � 𝑤𝑤 1 + , … , 𝑤𝑤 𝑖𝑖 + , … , 𝑤𝑤 𝑛𝑛 + � , where 𝑤𝑤 𝑖𝑖 + = � max 𝑤𝑤 𝑖𝑖𝑖𝑖 if 𝑗𝑗 ∈ 𝐿𝐿𝐿𝐿 min 𝑤𝑤 𝑖𝑖𝑖𝑖 if 𝑗𝑗 ∈ 𝑃𝑃𝐿𝐿 � for 𝑗𝑗 = 1,2, … , 𝑛𝑛 (11) Begic-Hajdarevic, Klančnik, Muhamedagić, Čekić, Cohodar Husic, Ficko, Gusel 112 Advances in Production Engineering & Management 19(1) 2024 𝑁𝑁𝑃𝑃 = � 𝑤𝑤 1 − , … , 𝑤𝑤 𝑖𝑖 − , … , 𝑤𝑤 𝑛𝑛 − � , where 𝑤𝑤 𝑖𝑖 − = � min 𝑤𝑤 𝑖𝑖𝑖𝑖 if 𝑗𝑗 ∈ 𝐿𝐿𝐿𝐿 max 𝑤𝑤 𝑖𝑖𝑖𝑖 if 𝑗𝑗 ∈ 𝑃𝑃𝐿𝐿 � for 𝑗𝑗 = 1,2, … , 𝑛𝑛 (12) Step 11: Obtain the distances of each experimental trials in relation to ideal solutions: 𝑃𝑃 𝑖𝑖 + = � � � 𝑤𝑤 𝑖𝑖𝑖𝑖 − 𝑤𝑤 𝑖𝑖 + � 2 𝑛𝑛 𝑖𝑖 = 1 , 𝑖𝑖 = 1,2, … , 𝑚𝑚 (13) 𝑃𝑃 𝑖𝑖 − = � � � 𝑤𝑤 𝑖𝑖𝑖𝑖 − 𝑤𝑤 𝑖𝑖 − � 2 𝑛𝑛 𝑖𝑖 = 1 , 𝑖𝑖 = 1,2, … , 𝑚𝑚 (14) Step 12: Calculate the closeness index value: 𝐶𝐶 𝑖𝑖 = 𝑃𝑃 𝑖𝑖 − 𝑃𝑃 𝑖𝑖 + + 𝑃𝑃 𝑖𝑖 − , 𝑖𝑖 = 1, 2, … , 𝑚𝑚 (15) Step 13: Rank the closeness index in the descending order. 3. Results and discussion In order to demonstrate and prove of the effectiveness of the proposed PSI-TOPIS method, prac- tical example of the selection of FDM process parameters were presented. Also, the results ob- tained by the proposed hybrid method were compared with the results obtained using other MCDM methods. The experimental data from Table 3 were normalized using Eq. 4 and the matrix was shown in Table 4. In Table 5 were presented data that were calculated using Eq. 6, 7 and 8, as well as the weight factors for each criterion (using Eq. 9). From Table 5 it can be seen that the compres- sive strength is most important criteria. The weighted normalized decision matrix was deter- mined using Eq. 10 and this matrix is also shown in Table 4 due to space limitation. Using Eqs. 11 and 12, the positive and negative ideal solution were determined. Further, the distances of each experimental trials (alternatives) in relation to positive and negative ideal so- lution were calculated using Eq. 13 and 14 and given in Table 6. Also, Table 6 shows the close- ness index calculated using Eq. 15 and the ranking order of given alternatives. Table 4 Matrix 𝑁𝑁 𝑖𝑖𝑖𝑖 and 𝑤𝑤 𝑖𝑖𝑖𝑖 Exp. No. Normalized matrix Weighted normalized matrix FS (MPa) FM (MPa) CS (MPa) FS (MPa) FM (MPa) CS (MPa) 1 0.4179 0.8311 0.9381 0.0912 0.3059 0.3880 2 0.8842 1.0000 0.9717 0.1930 0.3681 0.4019 3 1.0000 0.9794 1.0000 0.2183 0.3605 0.4136 4 0.3960 0.5952 0.8195 0.0865 0.2191 0.3389 5 0.8648 0.8618 0.9418 0.1888 0.3172 0.3896 6 0.7832 0.8790 0.9217 0.1710 0.3236 0.3812 7 0.3214 0.6815 0.8270 0.0702 0.2508 0.3420 8 0.5648 0.7369 0.8648 0.1233 0.2713 0.3577 9 0.7847 0.8301 0.8579 0.1713 0.3055 0.3548 Table 5 Data determination using Eqs. 6-9 Criteria 𝑁𝑁 𝑖𝑖 𝜑𝜑 𝑖𝑖 ∆ 𝑖𝑖 𝑝𝑝 𝑖𝑖 FS 0.6686 0.4898 0.5102 0.2183 FM 0.8217 0.1399 0.8601 0.3681 CS 0.9047 0.0334 0.9666 0.4136 FDM process parameter selection by hybrid MCDM approach for flexural and compression strength maximization Advances in Production Engineering & Management 19(1) 2024 113 Table 6 Closeness index and ranking Exp. No. S 𝑖𝑖 + S 𝑖𝑖 − 𝐶𝐶 𝑖𝑖 Rank 1 0.1438 0.1019 0.4149 6 2 0.0278 0.2031 0.8794 2 3 0.0076 0.2180 0.9664 1 4 0.2125 0.0163 0.0712 9 5 0.0635 0.1621 0.7183 3 6 0.0726 0.1512 0.6756 4 7 0.2020 0.0319 0.1365 8 8 0.1467 0.0768 0.3436 7 9 0.0979 0.1340 0.5779 5 Results from Table 6, that were obtained using hybrid PSI-TOPSIS method, show that the al- ternative 𝐸𝐸 3 is the best option, while the alternative 𝐸𝐸 4 is the worst choice. Flexural and com- pressive stress-strain curves for the best and worst alternatives are shown in Fig. 1. The optimal combination of FDM input parameters for printing PLA parts with regard to the considered pro- cess performance are 220°C printing temperature, 0.10 mm layer thickness and 90° raster angle, as also shown in Table 7. In this table the bold value indicates level at optimal parameter set- tings for individual input parameters. It is clear that printing temperature has the most signifi- cant effect on the process performance, followed by layer thickness and then raster angle. Table 7 Response table for the mean 𝐶𝐶 𝑖𝑖 . Input parameters Closeness index max.-min. Rank Level 1 Level 2 Level 3 T 0.2075 0.6471 0.7400 0.5325 1 L 0.7536 0.4884 0.3527 0.4009 2 A 0.4780 0.5095 0.6071 0.1291 3 Fig. 1 The flexural (left) and compressive (right) stress-strain curves for the best (sample 3) and worst (sample 4) options The effect of input parameters on the flexural strength and compressive strength are illus- trated in Fig. 2 and Fig. 3, respectively. Results showed that flexural strength and compressive strength increased by increasing the printing temperature. This can be explained by the fact that by increasing the printing temperature, the stronger cohesive forces were realized between in- dividual raster and layers, that resulted in higher flexural and compressive strength. By increasing of layer thickness, flexural strength and compressive strength decrease, be- cause the porosity between individual layers increases. Also, it can be observed that the highest values of flexural and compressive strength were achieved at a raster angle of 90°, because in this case, the direction of material deposition coincides with the direction of the load. The lowest value of the flexural strength was obtained at a raster angle of 0°, because in this case the strength of the test samples primarily depends on the cohesive force between individual raster. While the lowest value of the compressive strength was achieved at a raster angle of 45° due to the shear stresses between individual raster. Begic-Hajdarevic, Klančnik, Muhamedagić, Čekić, Cohodar Husic, Ficko, Gusel 114 Advances in Production Engineering & Management 19(1) 2024 Fig. 2 Mean of the flexural strength for different levels of input parameters Fig. 3 Mean of the compressive strength for different levels of input parameters Therefore, based on everything stated above, it can be concluded that the proposed hybrid PSI-TOPSIS method provides effectively very good results. According to the proposed method, the best option (alternative no. 3) was achieved at the highest varied printing temperature, the smallest varied layer thickness and the raster angle of 90°. The best choice (alternative 3) was confirmed by all other MCDM methods (GRA, TOPSIS and TOPSIS-ENTROPY), as can be seen in Fig. 4. The worst option, as predicted by the hybrid PSI-TOPSIS method, is alternative no. 4. This was not predicted by the other MCDM methods, as clearly shown in Fig. 4. Thus, the worst choice, as predicted by the others considered methods, is alternative no.7. The worst option, predicted by the proposed PSI-TOPSIS method, was achieved at the lowest varied printing tem- perature, the mean value of the layer thickness and the raster angle of 45° (it is most unfavoura- ble angle for the compressive strength, as seen in Fig. 3). Alternative 7, as the worst choice pre- dicted by the other methods, was also achieved at the lowest value of the printing temperature. Given that the results (as seen in Table 7) showed that the printing temperature has a most im- portant effect on the process performance and that by decreasing the printing temperature the flexural and compressive strength decrease (as shown in Fig. 2 and Fig. 3), this proves the effec- tiveness of the proposed method. The effectiveness of the proposed method is also proven by the fact that the worst option (alternative 4) was achieved at the most unfavourable raster angle for the compressive strength (compressive strength is the most important criteria, as shown in Table 5). This was not predicted by the other MCDM methods. The worst option predicted by the 0.3 0.2 0.1 80 70 60 50 40 30 220 200 180 90 45 0 Layer Thickness Mean of Flexural Strength, MPa Printing Temperature Raster Angle 0.3 0.2 0.1 47 46 45 44 43 42 41 220 200 180 90 45 0 Layer Thickness Mean of Compressive Strength, MPa Printing Temperature Raster Angle FDM process parameter selection by hybrid MCDM approach for flexural and compression strength maximization Advances in Production Engineering & Management 19(1) 2024 115 other MCDM methods was achieved at the raster angle of 90°. This raster angle is the most fa- vourable angle for both considered criteria (flexural strength and compressive strength), as seen in Figs. 2 and 3. Thus, in this paper, the determination of the best option does not depend on the MCDM methods used, it was also shown in [20]. However, the worst alternative predicted by the pro- posed hybrid method, unlike the other methods used, shows a good ranking order of the alterna- tives by the proposed method, that is an advantage proposed method in compared to the other methods used. Certainly, this advantage offered by the proposed method should be proven in other cases, that is a suggestion for future research. Fig. 4 Comparison of ranking with different MCDM methods 4. Conclusion In this paper, a novel hybrid PSI-TOPSIS methodology was presented. The proposed method was tested on the example of selecting optimal process parameters during FDM printing of PLA sam- ples. Also, the results obtained by the PSI-TOPSIS method were compared with the results that obtained by other MCDM methods. The results show that a printing temperature of 220°C, a lay- er thickness of 0.10 mm and a raster angle of 90° would be the best choice of process parame- ters according to PSI-TOPSIS analysis which has the best combination of mechanical properties of the tested samples. In future research, a hybrid PSI-TOPSIS method will be proposed for the selection of process parameters in other non-conventional machining processes, such as laser cutting or abrasive water jet cutting. Acknowledgement The authors acknowledge the financial support from the Slovenian Research Agency (research core funding No. P2- 0157) and the Ministry of Science, Higher Education and Youth of the Sarajevo Canton. 1 2 3 4 5 6 7 8 9 hybrid PSI-TOPSIS 6 2 1 9 3 4 8 7 5 GRA 6 2 1 8 3 4 9 7 5 TOPSIS 7 2 1 8 3 4 9 6 5 TOPSIS-ENTROPY 7 2 1 8 3 5 9 6 4 0 1 2 3 4 5 6 7 8 9 10 Rank Alternative hybrid PSI-TOPSIS GRA TOPSIS TOPSIS-ENTROPY Begic-Hajdarevic, Klančnik, Muhamedagić, Čekić, Cohodar Husic, Ficko, Gusel 116 Advances in Production Engineering & Management 19(1) 2024 References [1] Shanmugam, V., Pavan, M.V., Babu, K., Karnan, B. (2021). Fused deposition modeling based polymeric materials and their performance: A review, Polymer Composites, Vol. 42, No. 11, 5656-5677, doi: 10.1002/pc.26275. [2] Tripathy, C.R., Sharma, R.K., Rattan, V.K. (2022). Effect of printing parameters on the mechanical behaviour of the thermoplastic polymer processed by FDM technique: A research review, Advances in Production Engineering & Management, Vol. 17, No. 3, 279-294, doi: 10.14743/apem2022.3.436. [3] Alafaghani, A., Qattawi, A. (2018). Investigating the effect of fused deposition modeling processing parameters using Taguchi design of experiment method, Journal of Manufacturing Processes, Vol. 36, 164-174, doi: 10.1016/j.jmapro.2018.09.025. [4] Shakeri, Z., Benfriha, K., Zirak, N., Shirinbayan, M. (2022). Mechanical strength and shape accuracy optimization of polyamide FFF parts using grey relational analysis, Scientific Reports, Vol. 12, No. 1, Article No. 13142, doi: 10.1038/s41598-022-17302-z. [5] Rajamani, D., Balasubramanian, E., Yang, L.-J. (2022). Enhancing the surface quality of FDM processed flapping wing micro mechanism assembly through RSM–TOPSIS hybrid approach, Processes, Vol. 10, No. 11, Article No. 2457, doi: 10.3390/pr10112457. [6] Muhamedagic, K., Berus, L., Potočnik, D., Cekic, A., Begic-Hajdarevic, D., Cohodar Husic, M., Ficko, M. (2022). Effect of process parameters on tensile strength of FDM printed carbon fiber reinforced polyamide parts, Applied Sciences, Vol. 12, No. 12, Article No. 6028, doi: 10.3390/app12126028. [7] Omer, R., Mali, H.S., Singh, S.K. (2020). Tensile performance of additively manufactured short carbon fibre-PLA composites: Neural networking and GA for prediction and optimisation, Plastics, Rubber and Composites: Macro- molecular Engineering, Vol. 49, No. 6, 271-280, doi: 10.1080/14658011.2020.1744371. [8] Yadav, D., Chhabra, D., Gupta, R.K., Phogat, A., Ahlawat, A. (2020). Modeling and analysis of significant process parameters of FDM 3D printer using ANFIS, Materials Today: Proceedings, Vol. 21, Part 3, 1592-1604, doi: 10.1016/j.matpr.2019.11.227. [9] Deshwal, S., Kumar, A., Chhabra, D. (2020). Exercising hybrid statistical tools GA-RSM, GA-ANN and GA-ANFIS to optimize FDM process parameters for tensile strength improvement, CIRP Journal of Manufacturing Science and Technology, Vol. 31, 189-199, doi: 10.1016/j.cirpj.2020.05.009. [10] Visagan, A., Ganesh, P. (2022). Parametric optimization of two point incremental forming using GRA and TOPSIS, International Journal of Simulation Modelling, Vol. 21, No. 4, 615-626, doi: 10.2507/IJSIMM21-4-622. [11] Ficko, M., Begic-Hajdarevic, D., Hadziabdic, V., Klancnik, S. (2020). Multi-response optimisation of turning pro- cess parameters with GRA and TOPSIS methods, International Journal of Simulation Modelling, Vol. 19, No. 4, 547-558, doi: 10.2507/IJSIMM19-4-524. [12] Patalas-Maliszewska, J., Losyk, H. (2022). An approach to maintenance sustainability level assessment integrated with Industry 4.0 technologies using Fuzzy-TOPSIS: A real case study, Advances in Production Engineering & Management, Vol. 17, No. 4, 455-468, doi: 10.14743/apem2022.4.448. [13] Srinivasan, R., Pridhar, T., Ramprasath, L.S., Sree Charan, N., Ruban, W. (2020). Prediction of tensile strength in FDM printed ABS parts using response surface methodology (RSM), Materials Today Proceedings, Vol. 27, Part 2, 1827-1832, doi: 10.1016/j.matpr.2020.03.788. [14] John, J., Devjani, D., Ali, S., Abdallah, S., Pervaiz, S. (2023). Optimization of 3D printed polylactic acid structures with different infill patterns using Taguchi-grey relational analysis, Advanced Industrial and Engineering Polymer Research, Vol. 6, No. 1, 62-78, doi: 10.1016/j.aiepr.2022.06.002. [15] Wankhede, V., Jagetiya, D., Joshi, A., Chaudhari, R. (2020). Experimental investigation of FDM process parameters using Taguchi analysis, Materials Today: Proceedings, Vol. 27, Part 3, 2117-2120, doi: 10.1016/j.matpr.2019.09. 078. [16] Chohan, J.S., Kumar, R., Singh, T.B., Singh, S., Sharma, S., Singh, J., Mia, M., Pimenov, D.Y., Chattopadhyaya, S., Dwivedi, S.P., Kapłonek, W. (2020). Taguchi S/N and TOPSIS based optimization of fused deposition modelling and vapor finishing process for manufacturing of ABS plastic parts, Materials, Vol. 13, No. 22, Article No. 5176, doi: 10.3390/ma13225176. [17] Patel, P.B., Patel, J.D., Maniya, K.D. (2018). Application of PSI methods to select FDM process parameter for pol- ylactic acid, Materials Today: Proceedings, Vol. 5, No. 2, 4022-4028, doi: 10.1016/j.matpr.2017.11.662. [18] Madić, M., Antucheviciene, J., Radovanović, M., Petković, D. (2017). Determination of laser cutting process condi- tions using the preference selection index method, Optics & Laser Technology, Vol. 89, 214-220, doi: 10.1016/j.optlastec.2016.10.005. [19] Yadav, S., Pathak, V.K., Gangwar, S. (2019). A novel hybrid TOPSIS-PSI approach for material selection in marine applications, S ādh an ā, Vol. 44, Article No. 58, doi: 10.1007/s12046-018-1020-x. [20] Trung, D.D., Thinh, H.X. (2021). A multi-criteria decision-making in turning process using the MAIRCA, EAMR, MARCOS and TOPSIS methods: A comparative study, Advances in Production Engineering & Management, Vol. 16, No. 4, 443-456, doi: 10.14743/apem2021.4.412.