Microsoft Word - ETASR_V12_N4_pp8896-8900


Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8896-8900 8896 
 

www.etasr.com Ternane et al.: Regression Modeling and Process Analysis of Resistance Spot Welding on Dissimilar … 

 

Regression Modeling and Process Analysis of 

Resistance Spot Welding on Dissimilar Steel Sheets 
 

Fouad Ternane 

Department of Mechanical Engineering 

IS2M Laboratory, University of Aboubekr Belkaid 

Tlemcen, Algeria 

fouadsciences@yahoo.fr 

Mustapha Benachour  

Department of Mechanical Engineering 

IS2M Laboratory, University of Aboubekr Belkaid 

Tlemcen, Algeria 

bmf_12002@yahoo.fr 

Fethi Sebaa 

Department of Mechanical Engineering 

IS2M Laboratory, University of Aboubekr Belkaid 

Tlemcen, Algeria 

sebaafethi@yahoo.fr 

Nadjia Benachour  

Department of Mechanical Engineering 

IS2M Laboratory, University of Aboubekr Belkaid 

Tlemcen, Algeria 

 

nbenachour2005@yahoo.fr 
 

Received: 10 May 2022 | Revised: 27 May 2022 | Accepted: 28 May 2022 

 

Abstract-The resistance spot welding process is used to weld 

dissimilar materials. Dissimilar joining is formed by two 2mm 

thick sheet metals of 304L austenitic stainless steel and 

galvanized steel. This study investigates the effects of welding 

parameters such as welding current, welding time, and welding 

force. The welding time and the welding force were, respectively, 

in the range of 10-13 cycles and 7-8bar, while the welding current 

was in the range of 10–16kA. Tensile tests were applied to 

determine the resistance parameters of dissimilar joining. The 

experimental results showed that increasing the welding current 

increased the tensile shear stress of the weld coupon. Regression 

analysis was carried out to determine the significance of the 

process parameters by using the coupled of the full factorial 

experimental design with statistical and graphical analysis of the 

results. Furthermore, analysis of variance was used to determine 

the optimal parameters and combinations to achieve the highest 

strength level. 

Keywords-resistance spot welding; dissimilar joining; stainless 

steel & galvanized steel; welding parameters; regression model; 

tensile shear stress 

I. INTRODUCTION  

Resistance Spot Welding (RSW) is the main welding 
process used in the manufacturing industry, especially in the 
manufacturing of automobile truck cabins, rail vehicles, 
motorcycles, and home appliances such as refrigerators. For 
example, there are 3000 to 12000 spot welds in a car [1, 2]. The 
majority of the research investigations in spot welding has been 
carried out on the joining of homogeneous materials [3-6]. 
Currently, research in different industries is directed towards 
the reduction of CO2 emissions. The combination of 
heterogeneous materials is necessary to reduce the weight of 
the structures and CO2 emissions [7-9]. Engineers are 
increasingly looking to join different material types ensuring 

good mechanical strength, corrosion resistance, and sealing 
function [10, 11]. RSW of dissimilar materials was investigated 
in [10], finding that the lap shear strength depends on the 
strength and thickness of the non-stainless steels. A failure in 
dissimilar joining is called a plug failure. 

The properties of RSW of dissimilar stainless steel lap 
joints were studied in [12], where the weld quality of dissimilar 
joints was linked to their tensile-shear strength. The welding 
current was the main governing factor that affected the tensile-
shear strength of the RSW specimens. The current contribution 
of the weld was approximately 67% compared to weld time 
and applied load. As the weld current increases, the size of the 
weld nugget also increases. The mechanical performance of 
dissimilar stainless steel and carbon steel depends on the size of 
the fusion zone on the carbon steel side, where the pullout 
failure location of the carbon and stainless steel joint was in the 
Base Metal (BM) of the carbon steel side [11]. Tensile-shear 
properties and types of fracture of RSW joints were examined 
in [1], showing that increasing the weld time increased the 
tensile shear load-bearing capacity. Pullout and tearing failures 
are identified depending on the dissimilar joints (back 
hardening steel/304 stainless steel and back hardening 
steel/Interstitial steel). Pullout and interfacial failure modes 
were observed in several studies of homogeneous and 
inhomogeneous RSW of stainless steel and low carbon steel 
[11, 13-18]. The effects of welding current and welding process 
on residual stress distributions for dissimilar welded joints 
between stainless steel 316L and ferritic steel were investigated 
in [19], which have been used for lifetime assessment and 
failure mode predictions of the welded joints.  

Several studies have been conducted to optimize the 
welding parameters [22-24]. Factorial analysis was applied to 
investigate the effect of feed water temperature, flow rate, and 

Corresponding author: Fouad Ternane



Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8896-8900 8897 
 

www.etasr.com Ternane et al.: Regression Modeling and Process Analysis of Resistance Spot Welding on Dissimilar … 

 

radiator dimension on the total heat transfer coefficient and 
entropy generation in the panel radiator [22]. The effect of 
welding parameters on the tensile strength of RSW of the 
5052Al alloy using a factorial design was studied in [25]. The 
effects of RSW process parameters and their interactions were 
studied in [26] using a non-linear regression model. 

This paper presents an experimental study on determining 
the effect of the RSW parameters on the tensile-shear strength 
of dissimilar joints of stainless steel 304L with galvanized steel 
Z275. Linear regression, variance analysis, and a mathematical 
model were applied to evaluate the effects of the RSW 
parameters and some combinations on tensile-shear stress. 

II. EXPERIMENTAL PROCEDURES 

This study used 304L Stainless Steel (SS) and Z275 
Galvanized Steel (GS), each having a thickness of 2mm. The 
chemical compositions of the steels are given in Table I. The 
dimensions of the tensile specimens were 85mm length and 
30mm width, as shown in Figure 1. The total length of the 
RSW welded specimen was 140 mm. 

TABLE I.  CHEMICAL COMPOSITION OF THE WELDING MATERIALS 

Element C Si Mn P S Cr Ni 

SS 304L 0.03 1.0 2.0 0.045 0.015 0.1 0.03 

GS Z275 0.12 0.50 0.60 0.10 0.045 / / 

 

 

Fig. 1.  Dimensions of the tensile test specimens. 

The samples were welded by a TECNA ART 8201N RSW 
machine. Welding was performed using a water-cooled 
electrode with a hemispherical active face. The electrode in the 
CuNi2Be alloy had 6.0mm in contact surface according to: 

� ������ 	 2� � 3    (1) 

where D is the active diameter and e is the sheet thickness in 
mm. 

III. WELDING PARAMETERS DETERMINATION 

The welding parameters were determined from previous 
research and the weldability of materials, depending on the 
nature of the base materials and thickness [20]. The welding 
current varied from 10kA to 16kA, the welding time increased 
from 10 to 13 cycles, and two values of the electrode force (7 
and 8bar) were also applied. The distribution of the values of 
the three parameters was carried out with a factorial plan to 
ensure the union of all the values. Table II shows the 
combination of the welding parameters and the number of trials 
for each factor. The mechanical properties of joining SS 304L 
and GS Z275 materials are given in Table III. 

TABLE II.  FACTORS OF WELDING PARAMETERS 

Factor Levels Values 

I (kA) 5 10; 12; 14; 15; 16 

T (cycles) 3 10; 11; 13 

F (bar) 2 7; 8 

TABLE III.  MECHANICAL PROPERTIES OF 304L SL AND Z275 GS 

Materials E (GPa) 
��MPa) UTS (MPa) A% HRB 
SS 304 L 190 336 655 43.2 60 

GS Z275 200 266 330 22.0 / 

 

IV. RESULTS AND DISCUSSION 

A linear regression model was proposed to relate the 
tensile-shear strength of the RSW joints to welding current I, 
welding time T, and welding force F. Minitab 19 was used for 
regression analysis. As a result, a mathematical model was 
developed for the influence of the three welding parameters on 
tensile-shear stress, described as: 

� 	  � �  �1�1 �  �2�2 �  �3�3 �  �11�1�1  �
 �23�2�3    (2) 

The regression equation between the welding parameters I, 
T, and F and tensile-shear strength was: 

������� �ℎ��� ������ ����� 	 �343 �  33,2 ! �
 35,4 � �  47,4 $ �  0,767 !² �  4,58 � ∗ $    (3) 

Table IV shows the results of tensile-shear stress with the 
parameters of each test. 

TABLE IV.  EXPERIMENTAL DATA FOR TENSILE-SHEAR STRESS 

Trial order I (kA) T (Cycles) F (bar) Tensile shear stress (MPa) 

1 10 10 7 292.01 

2 10 10 8 265.07 

3 10 11 7 279.15 

4 10 11 8 271.81 

5 10 13 7 303.64 

6 10 13 8 276.71 

7 12 10 7 287.72 

8 12 10 8 320.17 

9 12 11 7 309.15 

10 12 11 8 319.56 

11 12 13 7 314.66 

12 12 13 8 280.99 

13 14 10 7 348.94 

14 14 10 8 342.21 

15 14 11 7 335.48 

16 14 11 8 331.19 

17 14 13 7 346.49 

18 14 13 8 352.00 

19 15 10 7 348.33 

20 15 10 8 355.07 

21 15 11 7 353.23 

22 15 11 8 350.78 

23 15 13 7 355.07 

24 15 13 8 350.17 

25 16 10 7 363.02 

26 16 10 8 355.07 

27 16 11 7 344.66 

28 16 11 8 348.33 

29 16 13 7 363.02 

30 16 13 8 356.90 

 



Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8896-8900 8898 
 

www.etasr.com Ternane et al.: Regression Modeling and Process Analysis of Resistance Spot Welding on Dissimilar … 

 

Table V illustrates the results of the analysis of variance 
with some interactions between the three welding parameters. 
The data show that the main effects of the welding current are 
relevant for the maximum load. Since the p-value of welding 
current is inferior to the significance level established at a 5% 
probability level (p<0.05), the interaction between welding 
time and electrode force cannot be considered significant. This 
Table also shows other adequacy measures: R²=88.96%, 
adjusted R²=86.66%, and predicted R²=81.59%. Since they are 
all very close to 0.9, they indicate a suitable model. 

TABLE V.  ANALYSIS OF VARIANCE 

Source DF Adj. SS Adj. MS F.Value P.Value 

Regression 5 25237.7 5047.5 38.68 0.000 

I 1 632.6 632.6 4.85 0.038 

T 1 258.8 258.8 1.98 0.172 

F 1 201.3 201.3 1.54 0.226 

I*I 1 228.4 228.4 1.75 0.198 

T*F 1 245.0 245.0 1.88 0.183 

Error 24 3132.0 130.5   

Total 29 28369.7    

R²= 88.96% 

Adjusted R² = 86.66% 

Predicted R² = 81.59% 
 

It is necessary to compare the tensile shear stress values 
estimated by the model equations (2), (3) with the values 
obtained as a result of experimental calculations. The residuals 
consist of the difference between the data and the model data. 
A comparison of the estimated and experimental data for 
tensile shear stress is shown in Figure 2 [26, 27]. According to 
Figure 2, a good agreement was observed between the 
mathematical model obtained for the tensile-shear stress and 
the experimental data, and residuals fall relatively along a 
straight line. Consequently, the normal distribution assumption 
was considered satisfied. 

 

 

Fig. 2.  Normal probability plot of residuals. 

Figure 3 is a residual versus observation order graph 
showing that all residual points are spread within the lower and 
upper bounds without evident patterns, confirming the 
assumption that the residuals have a regular variance. As a 
result, all diagnostic plots denote that all the necessary 
ANOVA assumptions are satisfied. 

 

Fig. 3.  Residuals versus observation order. 

As shown in Figure 4, all residual points are dispersed 
within the lower and upper bounds, showing no pattern. This 
graph denotes that the independence assumption is also 
satisfied. The histogram shown in Figure 5 forms a normal 
curve equally distributed around zero, showing that the 
normality assumption is more than likely true. 

 

 
Fig. 4.  Residuals versus fits plots. 

 

Fig. 5.  Histogram of residuals. 

As shown in Figure 6, a Pareto diagram provides statistical 
information on the effects of input variables, factors or welding 
parameters in this case, on tensile shear stress. It is observed 
that the welding current factor has a relevant impact on the 



Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8896-8900 8899 
 

www.etasr.com Ternane et al.: Regression Modeling and Process Analysis of Resistance Spot Welding on Dissimilar … 

 

tensile-shear stress because it occurs outside the dotted line at 
2.064. The other parameters and interactions have a 
meaningless or less impact relative to the welding current. It is 
important to know the effect levels of welding parameters on 
tensile-shear stress. Using this information, one could choose 
which parameter is more important for perfect welding joining 
[22, 28]. 

 

 

Fig. 6.  Pareto chart of normalized effects. 

These factors can be classified in decreasing order: welding 
time, welding time electrode force interaction, welding current, 
welding current interaction, and welding force. This result can 
be further supported by taking into account the main effects 
and interaction graphs, as shown in Figures 7 and 8 
respectively. Figure 7 shows a graphical representation of the 
primary effects of the factors examined for the spot weld 
regarding tensile-shear stress. According to the graph, it can be 
concluded that the impact of a factor is directly linked to the 
slope and length of the line in the graphic. The greater the slope 
is, the higher the influence on the average maximum load 
increase will be when varying levels from low to high. 
According to Figure 7, the welding current has a significant 
impact on tensile-shear stress due to the higher slope, while 
welding time is less sensitive to the variability in tensile-shear 
stress compared to welding current. On the other hand, welding 
force has less effect on tensile shear stress. 

 

 

Fig. 7.  Main effects plot for tensile shear stress (MPa). 

Figure 8 shows the main effect graph for tensile-shear 
stress. The three two-factor interaction graphs denote a 
powerful interaction between welding current and welding 
time, and welding current and welding force. Tensile-shear 
stress reaches its highest when welding current and welding 
time are kept at a high level, but welding force is at a low level, 
16kA, 13cycles, and 7bar respectively. Similarly, tensile-shear 
stress reaches its minimum when welding current and welding 
time are both at low levels, but electrode force maintains a high 
level, i.e. 10kA, 10cycles, and 8bar respectively. 

 

 

Fig. 8.  Interaction plot for Tensile shear stress (MPa) 

V. CONCLUSION 

This paper presented a linear regression analysis used to 
investigate the effect of RSW parameters on the tensile-shear 
stress of dissimilar joints from experimental results. Based on 
this investigation, the following conclusions can be drawn: 

• The regression analysis showed that there is a linear 
relationship between welding parameters (welding current, 
welding time, and electrode force) and the tensile-shear 
strength of the RSW joints. 

• According to the p-values and the Pareto chart, the welding 
current had the highest impact on tensile-shear stress 
compared to welding time, welding force, and other 
combinations. 

• The welding process and the effects of parameters and 
interactions between them on tensile-shear strength can be 
analyzed based on regression models of the welding process 
on two heterogeneous sheets, stainless steel and galvanized 
steel, with a thickness of 2mm which can make it to be an 
assistant reference for the welding process. 

• The optimal parameters that gave a higher tensile-shear 
strength were higher welding current (16kA) and welding 
time (13cycles) and lower welding force (7bar). 

ACKNOWLEDGMENT 

The authors gratefully acknowledge the technical support 
and welding specimens from the SOREMEP Society in 
Tlemcen. 



Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8896-8900 8900 
 

www.etasr.com Ternane et al.: Regression Modeling and Process Analysis of Resistance Spot Welding on Dissimilar … 

 

REFERENCES 

[1] F. Hayat, "Resistance Spot Weldability of Dissimilar Materials: BH180-
AISI304L Steels and BH180-IF7123 Steels," Journal of Materials 
Science & Technology, vol. 27, no. 11, pp. 1047–1058, Nov. 2011, 
https://doi.org/10.1016/S1005-0302(11)60185-0. 

[2] M. R. A. Shawon, F. Gulshan, and A. S. W. Kurny, "Effect of Welding 
Current on the Structure and Properties of Resistance Spot Welded 
Dissimilar (Austenitic Stainless Steel and Low Carbon Steel) Metal 
Joints," Journal of The Institution of Engineers (India): Series D, vol. 
96, no. 1, pp. 29–36, Apr. 2015, https://doi.org/10.1007/s40033-014-
0060-6. 

[3] A. Deliou and B. Bouchouicha, "Fatigue crack propagation in welded 
joints X70," Frattura ed Integrità Strutturale, vol. 12, no. 46, pp. 306–
318, Sep. 2018, https://doi.org/10.3221/IGF-ESIS.46.28. 

[4] B. Ahmed, O. C. E. Bahri, B. Benattou, and T. Malika, "Numerical 
Simulation of a Steel Weld Joint and Fracture Mechanics Study of a 
Compact Tension Specimen for Zones of Weld Joint," Frattura ed 
Integrità Strutturale, vol. 13, no. 47, pp. 17–29, 2019, https://doi.org/ 
10.3221/IGF-ESIS.47.02. 

[5] M. N. James and M. Newby, "The interface between metallurgy and 
mechanics in material performance," Frattura ed Integrità Strutturale, 
vol. 4, no. 14, pp. 5–16, 2010, https://doi.org/10.3221/IGF-ESIS.14.01. 

[6] A. Alzahougi, M. Elitas, and B. Demir, "RSW Junctions of Advanced 
Automotive Sheet Steel by Using Different Electrode Pressures," 
Engineering, Technology & Applied Science Research, vol. 8, no. 5, pp. 
3492–3495, Oct. 2018, https://doi.org/10.48084/etasr.2342. 

[7] B. Mezrag, F. Deschaux-Beaume, and M. Benachour, "Control of mass 
and heat transfer for steel/aluminium joining using Cold Metal Transfer 
process," Science and Technology of Welding and Joining, vol. 20, no. 3, 
pp. 189–198, Mar. 2015, https://doi.org/10.1179/1362171814Y.0000 
000271. 

[8] A. O. İrizalp, H. Durmuş, N. Yüksel, and İ. Türkmen, "Cold metal 
transfer welding of AA1050 aluminum thin sheets," Matéria (Rio de 
Janeiro), vol. 21, pp. 615–622, Sep. 2016, https://doi.org/10.1590/ 
S1517-707620160003.0059. 

[9] A. Subrammanian, D. B. Jabaraj, and V. K. Bupesh Raja, "Investigation 
of Microstructure and Mechanical Properties of Resistance Spot Welded 
Dissimilar Joints Between Ferritic Stainless Steel and Weathering Steel," 
Applied Mechanics and Materials, vol. 766–767, pp. 770–779, 2015, 
https://doi.org/10.4028/www.scientific.net/AMM.766-767.770. 

[10] M. Alenius, P. Pohjanne, M. Somervuori, and H. Hänninen, "Exploring 
the mechanical properties of spot welded dissimilar joints for stainless 
and galvanized steels," Welding Journal, vol. 85, no. 12, pp. 305–313, 
2006. 

[11] M. Pouranvari and S. P. H. Marashi, "Similar and dissimilar RSW of 
low carbon and austenitic stainless steels: effect of weld microstructure 
and hardness profile on failure mode," Materials Science and 
Technology, vol. 25, no. 12, pp. 1411–1416, Dec. 2009, https://doi.org/ 
10.1179/026708309X12459430509292. 

[12] A. B. Verma, S. U. Ghunage, and B. B. Ahuja, "Resistance Welding of 
Austenitic Stainless Steels (AISI 304 with AISI 316)," presented at the 
5th International & 26th All India Manufacturing Technology, Design 
and Research Conference (AIMTDR 2014), Guwahati, India, Dec. 2014. 

[13] A. M. Pereira, J. M. Ferreira, A. Loureiro, J. D. M. Costa, and P. J. 
Bártolo, "Effect of process parameters on the strength of resistance spot 
welds in 6082-T6 aluminium alloy," Materials & Design (1980-2015), 
vol. 31, no. 5, pp. 2454–2463, May 2010, https://doi.org/10.1016/ 
j.matdes.2009.11.052. 

[14] M. Pouranvari, S. M. Mousavizadeh, S. P. H. Marashi, M. Goodarzi, and 
M. Ghorbani, "Influence of fusion zone size and failure mode on 
mechanical performance of dissimilar resistance spot welds of AISI 
1008 low carbon steel and DP600 advanced high strength steel," 
Materials & Design, vol. 32, no. 3, pp. 1390–1398, Mar. 2011, 
https://doi.org/10.1016/j.matdes.2010.09.010. 

[15] M. Elitas and B. Demir, "The Effects of the Welding Parameters on 
Tensile Properties of RSW Junctions of DP1000 Sheet Steel," 
Engineering, Technology & Applied Science Research, vol. 8, no. 4, pp. 
3116–3120, Aug. 2018, https://doi.org/10.48084/etasr.2115. 

[16] F. Nikoosohbat, S. Kheirandish, M. Goodarzi, M. Pouranvari, and S. P. 
H. Marashi, "Microstructure and failure behaviour of resistance spot 
welded DP980 dual phase steel," Materials Science and Technology, vol. 
26, no. 6, pp. 738–744, Jun. 2010, https://doi.org/10.1179/174328409 
X414995. 

[17] M. Ishak, L. H. Shah, I. S. R. Aisha, W. Hafizi, and M. R. Islam, "Study 
of Resistance Spot Welding Between AISI 301 Stainless Steel and AISI 
1020 Carbon Steel Dissimilar Alloys," Journal of Mechanical 
Engineering and Sciences, vol. 6, pp. 793–806, Jun. 2014, 
https://doi.org/10.15282/jmes.7.2014.7.0077. 

[18] M. Safari and H. Mostaan, "Dissimilar resistance spot welding of AISI 
1075 eutectoid steel to AISI 201 stainless steel," Journal of Advanced 
Materials and Processing, vol. 5, no. 1, pp. 44–56, Mar. 2017. 

[19] R. Konecná, L. Kunz, G. Nicoletto, and A. Baca, "Fatigue crack growth 
behavior of Inconel 718 produced by selective laser melting," Frattura 
ed Integrità Strutturale, vol. 10, no. 35, pp. 31–40, 2016, https://doi.org/ 
10.3221/IGF-ESIS.35.04. 

[20] M. Pouranvari, P. Marashi, and M. Goodarzi, "Failure Mode of 
Dissimilar Resistance Spot Welds Between Austenitic Stainless and Low 
Carbon Steels," in Proceedings of Metal 2008, Hradec nad Moravicí, 
Czech Republic, May 2008. 

[21] P. Russo Spena, M. De Maddis, F. Lombardi, and M. Rossini, 
"Dissimilar Resistance Spot Welding of Q&P and TWIP Steel Sheets," 
Materials and Manufacturing Processes, vol. 31, no. 3, pp. 291–299, 
Feb. 2016, https://doi.org/10.1080/10426914.2015.1048476. 

[22] K. Gelis, "Factorial experimental design for second law analysis of panel 
radiators as a function of radiator dimension," Journal of Building 
Engineering, vol. 43, Nov. 2021, Art. no. 102872, https://doi.org/ 
10.1016/j.jobe.2021.102872. 

[23] J. Valera, V. Miguel, A. Martínez, J. Naranjo, and M. Cañas, 
"Optimization of electrical parameters in Resistance Spot Welding of 
dissimilar joints of micro-alloyed steels TRIP sheets," Procedia 
Manufacturing, vol. 13, pp. 291–298, Jan. 2017, https://doi.org/ 
10.1016/j.promfg.2017.09.074. 

[24] V. C. Nguyen, T. D. Nguyen, and D. H. Tien, "Cutting Parameter 
Optimization in Finishing Milling of Ti-6Al-4V Titanium Alloy under 
MQL Condition using TOPSIS and ANOVA Analysis," Engineering, 
Technology & Applied Science Research, vol. 11, no. 1, pp. 6775–6780, 
Feb. 2021, https://doi.org/10.48084/etasr.4015. 

[25] P. Peasura, "Optimization of Resistance Spot Welding on Aluminum 
Magnesium 5052 Grade with 23 Factorial Designs," Advanced Materials 
Research, vol. 622–623, pp. 340–343, 2013, https://doi.org/10.4028/ 
www.scientific.net/AMR.622-623.340. 

[26] Y. Luo, J. Liu, H. Xu, C. Xiong, and L. Liu, "Regression modeling and 
process analysis of resistance spot welding on galvanized steel sheet," 
Materials & Design, vol. 30, no. 7, pp. 2547–2555, Aug. 2009, 
https://doi.org/10.1016/j.matdes.2008.09.031. 

[27] S. M. Hamidinejad, F. Kolahan, and A. H. Kokabi, "The modeling and 
process analysis of resistance spot welding on galvanized steel sheets 
used in car body manufacturing," Materials & Design, vol. 34, pp. 759–
767, Feb. 2012, https://doi.org/10.1016/j.matdes.2011.06.064. 

[28] T. Saeheaw, "Regression Modeling and Process Analysis of Plug and 
Spot Welds Used in Automotive Body Panel Assembly," International 
Journal of Engineering, vol. 33, no. 11, pp. 2384–2398, Nov. 2020, 
https://doi.org/10.5829/ije.2020.33.11b.29.