Research Article Multiattribute Assessment of Consumables...

10
Research Article Multiattribute Assessment of Consumables for TIG Welding of Aluminum Alloys P. K. Baghel and D. S. Nagesh Department of Mechanical Engineering, Delhi Technological University, Bawana Road, New Delhi 110042, India Correspondence should be addressed to P. K. Baghel; [email protected] Received 23 August 2016; Revised 18 October 2016; Accepted 20 October 2016 Academic Editor: Jean Yves Hascoet Copyright © 2016 P. K. Baghel and D. S. Nagesh. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Consumable materials for welding processes such as filler alloy, electrode, and shielding gas have major implications in welding technology. Selecting the appropriate set of consumable materials for welding aluminum alloys from various available alternates gives rise to a significant decision making problem. is study proposes a strategic justification tool based on integrated analytical hierarchy process, technique for order preference by similarity to ideal solution (AHP-TOPSIS), to screen and select the best possible combination of filler alloy, electrode, and shielding gas for TIG welding. Based on qualitative analysis of available alternates, ER5356 filler alloys and 0.8% zirconiated electrode are selected as suitable consumable in combination with pure argon as shielding gas. e proposed result is experimentally validated using mechanical property testing. 1. Introduction Dissimilar aluminum alloys combinations are widely used in aerospace and automobile applications such as inner body panel, heat shields, structural components, ladders, and space frames [1]. Several welding processes such as friction stir welding (FSW) and laser welding are gaining popularity for welding aluminum but TIG welding is still most widely used process for joining aluminum because of its heat control ability (narrow heat affected zone) and eco-friendly nature (no spatter and smoke). Although friction stir welding has several advantages like no voids and cracking, no distortion of workpiece, no need of filler material, no shielding gas required, clean and environment friendly process because there are no harmful effects like arc formation, radiation, release of toxic gas, and so forth, however as compared to TIG welding process some disadvantages of the friction stir welding process have been identified as well such that exit hole is leſt when the tool is withdrawn. is limitation is serious when pressure vessels and cargo oil tanks are manufactured. Heavy-duty clamping is required to hold the plates to combat large downward force. In addition to this it is less flexible that is difficult to weld when thickness varies. e ideal weld joint should have tensile strength close to base metal and must be free from hot cracks and residual stress. Consumable materials, namely, filler alloy, electrode, and shielding gas, play a vital role in deciding the manufac- turing quality and cost of weld joint [2]. When multiple choices are present, choosing the opti- mum material of desired properties from large number of available materials is a tedious task. It is essential to base the final decision using well-structured selection methodology in consideration with all possible relevant attributes or properties. Multiattribute decision making (MADM) models are generally used to provide an encompassing analysis of all involved multiple attributes. In an unstructured decision problem in welding pro- cesses, such models have recently been used by many researchers. Khorshidi et al. [3] used TOPSIS for selecting the optimal refinement condition to achieve maximum tensile property in Al-15% Mg 2 Si composite whereas Mirhedayatian et al. [4] employed TOPSIS approach and integrated fuzzy data envelopment analysis in welding process selection for repairing nodular cast iron engine block. e selection of consumable materials is of prime impor- tance to accomplish the joint of desired specifications; how- ever there are not enough investigations in this direction. Hindawi Publishing Corporation International Journal of Manufacturing Engineering Volume 2016, Article ID 7291898, 9 pages http://dx.doi.org/10.1155/2016/7291898

Transcript of Research Article Multiattribute Assessment of Consumables...

Research ArticleMultiattribute Assessment of Consumables forTIG Welding of Aluminum Alloys

P K Baghel and D S Nagesh

Department of Mechanical Engineering Delhi Technological University Bawana Road New Delhi 110042 India

Correspondence should be addressed to P K Baghel pushpbaghelpbgmailcom

Received 23 August 2016 Revised 18 October 2016 Accepted 20 October 2016

Academic Editor Jean Yves Hascoet

Copyright copy 2016 P K Baghel and D S NageshThis is an open access article distributed under theCreative CommonsAttributionLicense which permits unrestricted use distribution and reproduction in anymedium provided the originalwork is properly cited

Consumable materials for welding processes such as filler alloy electrode and shielding gas have major implications in weldingtechnology Selecting the appropriate set of consumable materials for welding aluminum alloys from various available alternatesgives rise to a significant decision making problem This study proposes a strategic justification tool based on integrated analyticalhierarchy process technique for order preference by similarity to ideal solution (AHP-TOPSIS) to screen and select the best possiblecombination of filler alloy electrode and shielding gas for TIGwelding Based on qualitative analysis of available alternates ER5356filler alloys and 08 zirconiated electrode are selected as suitable consumable in combination with pure argon as shielding gasThe proposed result is experimentally validated using mechanical property testing

1 Introduction

Dissimilar aluminum alloys combinations are widely used inaerospace and automobile applications such as inner bodypanel heat shields structural components ladders and spaceframes [1] Several welding processes such as friction stirwelding (FSW) and laser welding are gaining popularity forwelding aluminum but TIG welding is still most widely usedprocess for joining aluminum because of its heat controlability (narrow heat affected zone) and eco-friendly nature(no spatter and smoke) Although friction stir welding hasseveral advantages like no voids and cracking no distortionof workpiece no need of filler material no shielding gasrequired clean and environment friendly process becausethere are no harmful effects like arc formation radiationrelease of toxic gas and so forth however as comparedto TIG welding process some disadvantages of the frictionstir welding process have been identified as well such thatexit hole is left when the tool is withdrawn This limitationis serious when pressure vessels and cargo oil tanks aremanufactured Heavy-duty clamping is required to hold theplates to combat large downward force In addition to thisit is less flexible that is difficult to weld when thicknessvaries

The ideal weld joint should have tensile strength close tobase metal and must be free from hot cracks and residualstress Consumable materials namely filler alloy electrodeand shielding gas play a vital role in deciding the manufac-turing quality and cost of weld joint [2]

When multiple choices are present choosing the opti-mum material of desired properties from large number ofavailable materials is a tedious task It is essential to base thefinal decision using well-structured selection methodologyin consideration with all possible relevant attributes orproperties Multiattribute decision making (MADM)modelsare generally used to provide an encompassing analysis of allinvolved multiple attributes

In an unstructured decision problem in welding pro-cesses such models have recently been used by manyresearchers Khorshidi et al [3] used TOPSIS for selecting theoptimal refinement condition to achieve maximum tensileproperty in Al-15Mg2Si composite whereas Mirhedayatianet al [4] employed TOPSIS approach and integrated fuzzydata envelopment analysis in welding process selection forrepairing nodular cast iron engine block

The selection of consumable materials is of prime impor-tance to accomplish the joint of desired specifications how-ever there are not enough investigations in this direction

Hindawi Publishing CorporationInternational Journal of Manufacturing EngineeringVolume 2016 Article ID 7291898 9 pageshttpdxdoiorg10115520167291898

2 International Journal of Manufacturing Engineering

Table 1 List of attributes and alternates for filler alloy [1]

Properties ER5356 ER5556 ER5183 ER5654 ER5554(1) Yield strength (MPa) 240 158579 275 22063 12410(2) Tensile strength (MPa) 26889 31715 29647 22752 186(3) Elongation 2 17 14 17 17 8(4) Brinell hardness (HB) 105 65 99 83 39(load 500Kg thickness 10ohm)(5) Cost ($) 135 16 155 172 13(6) Longitudinal shear strength (MPa) 11721 13789 12755 8273 79289(7) Transverse shear strength (MPa) 17926 20684 19305 12410 10342(8) Impact strength (119869) 182 157 16 13 10

Table 2 List of attributes and alternates for electrode [2]

Properties Pure tungsten 2 thoriated 2 ceriated 15 lanthanated 08 zirconiated(1) Electrode oxide 001 26 32 25 03work function (eV)(2) Density (gcm3) 1925 1172 665 57 649(3) Volume of 2 6 38 52 57 69weight of electrode(4) Melting point (∘C) 3410 1750 795 826 1852(5) Electron negativity 17 13 10 11 12(according to pauling)(6) Cost ($) 2399 1499 184 2399 203(7) Particle size (Nm) 30 50 40 20 30

Therefore there is a need for an extensive selection frameworkbefore undertaking the manufacturing process

In present research technique for order preference bysimilarity to ideal solution (TOPSIS) integrated with analyti-cal hierarchy process (AHP) is employed to select an optimalcombination of consumable materials for welding dissimilaraluminum alloys

2 Materials and Methodology

Consumables in welding processes comprise filler alloy elec-trode material and shielding gas The filler alloy determinesthe transverse and longitudinal strength of the weld jointTherefore the filler material should have attributes like hightensile properties impact toughness and low cost There aremainly five filler alloys that are used and listed in Table 1

Table 2 presents the alternates and attributes of electrodeused in pulse TIG welding of aluminum alloys Tungstenelectrode used in TIG welding is a consumable It merelyserves as the terminal for the electric arc which producesthe heat needed to join the base metal being welded Theshape of the tungsten electrode tip is an important processvariable in GTA welding Tungsten electrode geometry hasan effect on the arc shape (thereby affecting the weld beadsize and shape) the weld penetration and point longevity ofthe electrode In addition to the arc shape different tungstenmaterials pose different characteristics which influence arcstart ability electrode life and contamination resistance

Selection of right shielding gas composition significantlyaffects metallurgical structure properties soundness and

heat treated response of weldThe desired characteristics withsuitable alternates are listed in Table 3

TOPSISmethod is employed to evaluate and find the bestsuitable alternative in the present work It was first proposedin 1981 [5]

The methodology involves the evaluation of Euclideandistance from the positive and the negative ideal solution forthe given alternateThebest possible alternativewill be closestto the positive ideal solution and far away from negative idealsolution

TOPSIS is a suitable choice for material selection asit is a relatively more systematic process It provides bothqualitative and quantitative analysis of the available dataTheTOPSIS method involves a tool for evaluation of weights ofeach attributecriterion In this study this task is performedusing AHPmethodThe relative weights hence generated areused in TOPSIS process to find the final priority ranking ofavailable alternatives

It was developed by Satty [6] to handle complex problemsbased on multiple criteria The flexibility and intuitiveness ofAHP method provide edge over other methods It facilitatesgroup decision making by finding the geometric mean of theindividual pairwise comparisons

Figure 1 depicts (AHP-TOPSIS)MADMmethodology forwelding process consumables selection

The AHP Method To determine the relative importance ofdifferent attributes with respect to the objective the followingsteps are followed

International Journal of Manufacturing Engineering 3

Table 3 List of attributes and alternates for shielding gas [7 8]

Properties Pure argon Pure helium 25 argon + 75helium 75 argon + 25 helium 50 argon + 50 helium

(1) Thermal conductivity 0016 01513 01174 00498 00836WmK at (1013 bar at 0∘C)(2) Gas density (lbwft) at 0∘C1 atm 01114 00111 00361 00863 00612

(3) Specific gravity (gcm3) air =l 138 01368 04476 10692 07584

(4) Ionization potential (eV) 157 245 223 179 201(5) Cost (per 100 g) ($) 05 52 4025 1675 415(6) Specific heat capacity at 0520 519 40225 16875 2855kJKgsdotK C119901 (1013 bar and 25∘C)

Table 4 Nine-point intensity of importance scale

Relative importance (119886119894119895) Description (119894 over 119895)1 Equal importance3 Moderate importance5 Strong importance7 Very strong importance9 Absolute importance2 4 6 8 Intermediate values

The multiple pairwise comparisons are based on a stan-dardized comparison scale of nine levels (Table 4)

Let 119862 = 119862119895 | 119895 = 1 2 119899 be the set of criteriaThe result of the pairwise comparison on 119899 criteria can be

summarized in an (119899times119899) evaluationmatrix119860 Every element119886119894119895 (119894 119895 = 1 2 119899) denotes the relative importance of factor119894 with respect to factor 119895 A criterion compared with itself isalways assigned the value 1 so themain diagonal entries of thepairwise comparison matrix are all 1

Square matrix119860 containing every element 119886119894119895 is shown in

119860 =( 1 11988612 sdot sdot sdot 119886111989811988621 1 sdot sdot sdot 1198862119898 sdot sdot sdot 1198861198981 1198861198982 sdot sdot sdot 1 ) 119886119894119895 = 1119886119895119894 119886119895119894 =0 (1)

(i) The relative normalized weight (119882119894) of each factor isevaluated by finding the geometric mean of 119894th rowand normalizing the geometric means of rows in thecomparison matrix

GM119894 = 1198861198941 times 1198861198942 times 1198861198943 times sdot sdot sdot times 1198861198941198951119899 119882119894 = GM119894sum119895=119899119895=1 GM119894 (2)

Matrix 119883 is obtained to denote an 119899-dimensional columnvector The sum of the weighted values is computed for theimportance degrees of alternatives then119883 = 119860 lowast119882 where119882 = [119882111988221198823 119882119899]119879 119883 = 119860 lowast119882 = [[[[[[[

1 11988612 sdot sdot sdot 119886111989911988621 1 sdot sdot sdot 1198862119899 sdot sdot sdot 1198861198991 1198861198992 sdot sdot sdot 1]]]]]]][[[[[[[11988211198822119882119899]]]]]]] =

[[[[[[[11986211198622119862119899]]]]]]]

(3)

(ii) The consistency values (CV) are calculated for thecluster of alternatives represented by the vector

CV119894 = 119862119894119882119894 (4)

(iii) The largest eigenvalue 120582max is obtained which is themean of the consistency values

(iv) The consistency index (CI) is calculated by

CI = 120582max minus 119899119899 minus 1 (5)

The AHP output quality is strictly related to theconsistency of the pairwise comparison judgements

(vi) The random index (RI) is obtained for the number offactors being used in decision making (Table 5)

(vii) The final consistency ratio (CR) is calculated as theratio of the CI and CR

CR = CIRI (6)

The acceptable upper limit for CR is 01 The evaluationprocess has to be repeated to improve the consistency if thefinal consistency ratio exceeds this value This consistencycheck is essential to establish the validity of the pairwisecomparison matrix evaluation

TOPSIS method consists of the following steps

4 International Journal of Manufacturing Engineering

Consumable material

(1) Filler alloy(2) Electrode(3) Shielding gas

Evaluation of attributes for each

material

AHP

Developing decision matrix

Evaluation and ranking of each

material

Decision of optimumconsumable material selction

Optimum filler alloy optimumelectrode optimum shielding gas

TOPSIS

Dissimilar aluminum alloy joint5083-O + 6061-T651

Applications Pressure vessels cargooil tanks gun mount base

Figure 1 Flow chart depicting integrated AHP-TOPSIS methodology for consumable materials selection

Table 5 Random index (RI) values

Criteria RI Criteria RI3 052 7 1354 089 8 1405 111 9 1456 125 10 149

Step 1 (construction of the normalized decision matrix)For the Euclidean length of a vector the element 119903119894119895 of the

normalized decisionmatrix119877 is evaluated using the followingtransformation

119903119894119895 = 119883119894119895radicsum119898119894=1 (119883119894119895)2 119895 = 1 2 119899 119894 = 1 2 119898 (7)

Here 119903119894119895 is the normalized preference measure of the 119894thalternative ldquo119898rdquo is the number of alternatives and ldquo119899rdquo is thenumber of criteria

International Journal of Manufacturing Engineering 5

Step 2 (construction of the weighted normalized decisionmatrix) Multiply the columns of the normalized decisionmatrix with the set of weights obtained by AHP119882 = (1199081 1199082 1199083 119908119899) to obtain weighted normalizeddecision matrix

119881 = 119877119882=(((((((

1199081 sdot 11990311 1199082 sdot 11990312 sdot sdot sdot 119908119899 sdot 11990311198991199081 sdot 11990321 1199082 sdot 11990322 sdot sdot sdot 119908119899 sdot 1199032119899 1199081 sdot 1199031198981 1199082 sdot 1199031198982 sdot sdot sdot 119908119899 sdot 119903119898119899)))))))

(8)

Step 3 (determination of the ideal and negative ideal solu-tions) The ideal solution and negative ideal solution valuesets are determined respectively as follows

119881+1 119881+2 119881+119899 = (max119894119881119894119895 | 119869 isin 119870) (min

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898 119881minus1 119881minus2 119881minus119899 = (min

119894119881119894119895 | 119869 isin 119870) (max

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898

(9)

where 119870 = 119895 = 1 2 3 119899119895 is associated with benefit criteria 1198701015840 = 119895 = 1 2 3 119899119895 is associated with cost criteria (10)

Step 4 (measurement of separation distances from idealand negative ideal solutions) Euclidean distances for eachalternative are respectively calculated as

119878+119894 = 119899sum119895=1 (119881119894119895 minus 119881+119895 )212 119894 = 1 2 119898

119878minus119894 = 119899sum119895=1 (119881119894119895 minus 119881minus119895 )212 119894 = 1 2 119898 (11)

Step 5 (calculation of the relative closeness to the idealsolution) The relative closeness to the ideal solution can bedefined as119862119894 = 119878minus119894119878+119894 + 119878minus119894 119894 = 1 2 119898 0 le 119862119894 le 1 (12)

The higher the closeness means the better the rank

Step 6 (ranking of the preference order) The preferenceorder is ranked on the basis of the order of 119862119894 Hence theoptimum alternative is the one which is nearer to the idealsolution and farther to the negative ideal solution

3 Result and Discussion

31 Computational Results The normalized decision matrixfor filler alloy electrode and shielding gas is calculated using(7) as shown in matrices (13) (14) and (15) respectively

[[[[[[[[[[[[[[[[

05351 04262 05064 05740 03994 04702 04821 0540103156 05632 04170 03553 04734 05531 05562 0465905900 04884 05064 05412 04586 05116 05191 0474804528 03918 05064 04538 05089 03318 03337 0385802470 03303 02383 02132 03846 03180 02781 02967

]]]]]]]]]]]]]]]] (13)

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

2 International Journal of Manufacturing Engineering

Table 1 List of attributes and alternates for filler alloy [1]

Properties ER5356 ER5556 ER5183 ER5654 ER5554(1) Yield strength (MPa) 240 158579 275 22063 12410(2) Tensile strength (MPa) 26889 31715 29647 22752 186(3) Elongation 2 17 14 17 17 8(4) Brinell hardness (HB) 105 65 99 83 39(load 500Kg thickness 10ohm)(5) Cost ($) 135 16 155 172 13(6) Longitudinal shear strength (MPa) 11721 13789 12755 8273 79289(7) Transverse shear strength (MPa) 17926 20684 19305 12410 10342(8) Impact strength (119869) 182 157 16 13 10

Table 2 List of attributes and alternates for electrode [2]

Properties Pure tungsten 2 thoriated 2 ceriated 15 lanthanated 08 zirconiated(1) Electrode oxide 001 26 32 25 03work function (eV)(2) Density (gcm3) 1925 1172 665 57 649(3) Volume of 2 6 38 52 57 69weight of electrode(4) Melting point (∘C) 3410 1750 795 826 1852(5) Electron negativity 17 13 10 11 12(according to pauling)(6) Cost ($) 2399 1499 184 2399 203(7) Particle size (Nm) 30 50 40 20 30

Therefore there is a need for an extensive selection frameworkbefore undertaking the manufacturing process

In present research technique for order preference bysimilarity to ideal solution (TOPSIS) integrated with analyti-cal hierarchy process (AHP) is employed to select an optimalcombination of consumable materials for welding dissimilaraluminum alloys

2 Materials and Methodology

Consumables in welding processes comprise filler alloy elec-trode material and shielding gas The filler alloy determinesthe transverse and longitudinal strength of the weld jointTherefore the filler material should have attributes like hightensile properties impact toughness and low cost There aremainly five filler alloys that are used and listed in Table 1

Table 2 presents the alternates and attributes of electrodeused in pulse TIG welding of aluminum alloys Tungstenelectrode used in TIG welding is a consumable It merelyserves as the terminal for the electric arc which producesthe heat needed to join the base metal being welded Theshape of the tungsten electrode tip is an important processvariable in GTA welding Tungsten electrode geometry hasan effect on the arc shape (thereby affecting the weld beadsize and shape) the weld penetration and point longevity ofthe electrode In addition to the arc shape different tungstenmaterials pose different characteristics which influence arcstart ability electrode life and contamination resistance

Selection of right shielding gas composition significantlyaffects metallurgical structure properties soundness and

heat treated response of weldThe desired characteristics withsuitable alternates are listed in Table 3

TOPSISmethod is employed to evaluate and find the bestsuitable alternative in the present work It was first proposedin 1981 [5]

The methodology involves the evaluation of Euclideandistance from the positive and the negative ideal solution forthe given alternateThebest possible alternativewill be closestto the positive ideal solution and far away from negative idealsolution

TOPSIS is a suitable choice for material selection asit is a relatively more systematic process It provides bothqualitative and quantitative analysis of the available dataTheTOPSIS method involves a tool for evaluation of weights ofeach attributecriterion In this study this task is performedusing AHPmethodThe relative weights hence generated areused in TOPSIS process to find the final priority ranking ofavailable alternatives

It was developed by Satty [6] to handle complex problemsbased on multiple criteria The flexibility and intuitiveness ofAHP method provide edge over other methods It facilitatesgroup decision making by finding the geometric mean of theindividual pairwise comparisons

Figure 1 depicts (AHP-TOPSIS)MADMmethodology forwelding process consumables selection

The AHP Method To determine the relative importance ofdifferent attributes with respect to the objective the followingsteps are followed

International Journal of Manufacturing Engineering 3

Table 3 List of attributes and alternates for shielding gas [7 8]

Properties Pure argon Pure helium 25 argon + 75helium 75 argon + 25 helium 50 argon + 50 helium

(1) Thermal conductivity 0016 01513 01174 00498 00836WmK at (1013 bar at 0∘C)(2) Gas density (lbwft) at 0∘C1 atm 01114 00111 00361 00863 00612

(3) Specific gravity (gcm3) air =l 138 01368 04476 10692 07584

(4) Ionization potential (eV) 157 245 223 179 201(5) Cost (per 100 g) ($) 05 52 4025 1675 415(6) Specific heat capacity at 0520 519 40225 16875 2855kJKgsdotK C119901 (1013 bar and 25∘C)

Table 4 Nine-point intensity of importance scale

Relative importance (119886119894119895) Description (119894 over 119895)1 Equal importance3 Moderate importance5 Strong importance7 Very strong importance9 Absolute importance2 4 6 8 Intermediate values

The multiple pairwise comparisons are based on a stan-dardized comparison scale of nine levels (Table 4)

Let 119862 = 119862119895 | 119895 = 1 2 119899 be the set of criteriaThe result of the pairwise comparison on 119899 criteria can be

summarized in an (119899times119899) evaluationmatrix119860 Every element119886119894119895 (119894 119895 = 1 2 119899) denotes the relative importance of factor119894 with respect to factor 119895 A criterion compared with itself isalways assigned the value 1 so themain diagonal entries of thepairwise comparison matrix are all 1

Square matrix119860 containing every element 119886119894119895 is shown in

119860 =( 1 11988612 sdot sdot sdot 119886111989811988621 1 sdot sdot sdot 1198862119898 sdot sdot sdot 1198861198981 1198861198982 sdot sdot sdot 1 ) 119886119894119895 = 1119886119895119894 119886119895119894 =0 (1)

(i) The relative normalized weight (119882119894) of each factor isevaluated by finding the geometric mean of 119894th rowand normalizing the geometric means of rows in thecomparison matrix

GM119894 = 1198861198941 times 1198861198942 times 1198861198943 times sdot sdot sdot times 1198861198941198951119899 119882119894 = GM119894sum119895=119899119895=1 GM119894 (2)

Matrix 119883 is obtained to denote an 119899-dimensional columnvector The sum of the weighted values is computed for theimportance degrees of alternatives then119883 = 119860 lowast119882 where119882 = [119882111988221198823 119882119899]119879 119883 = 119860 lowast119882 = [[[[[[[

1 11988612 sdot sdot sdot 119886111989911988621 1 sdot sdot sdot 1198862119899 sdot sdot sdot 1198861198991 1198861198992 sdot sdot sdot 1]]]]]]][[[[[[[11988211198822119882119899]]]]]]] =

[[[[[[[11986211198622119862119899]]]]]]]

(3)

(ii) The consistency values (CV) are calculated for thecluster of alternatives represented by the vector

CV119894 = 119862119894119882119894 (4)

(iii) The largest eigenvalue 120582max is obtained which is themean of the consistency values

(iv) The consistency index (CI) is calculated by

CI = 120582max minus 119899119899 minus 1 (5)

The AHP output quality is strictly related to theconsistency of the pairwise comparison judgements

(vi) The random index (RI) is obtained for the number offactors being used in decision making (Table 5)

(vii) The final consistency ratio (CR) is calculated as theratio of the CI and CR

CR = CIRI (6)

The acceptable upper limit for CR is 01 The evaluationprocess has to be repeated to improve the consistency if thefinal consistency ratio exceeds this value This consistencycheck is essential to establish the validity of the pairwisecomparison matrix evaluation

TOPSIS method consists of the following steps

4 International Journal of Manufacturing Engineering

Consumable material

(1) Filler alloy(2) Electrode(3) Shielding gas

Evaluation of attributes for each

material

AHP

Developing decision matrix

Evaluation and ranking of each

material

Decision of optimumconsumable material selction

Optimum filler alloy optimumelectrode optimum shielding gas

TOPSIS

Dissimilar aluminum alloy joint5083-O + 6061-T651

Applications Pressure vessels cargooil tanks gun mount base

Figure 1 Flow chart depicting integrated AHP-TOPSIS methodology for consumable materials selection

Table 5 Random index (RI) values

Criteria RI Criteria RI3 052 7 1354 089 8 1405 111 9 1456 125 10 149

Step 1 (construction of the normalized decision matrix)For the Euclidean length of a vector the element 119903119894119895 of the

normalized decisionmatrix119877 is evaluated using the followingtransformation

119903119894119895 = 119883119894119895radicsum119898119894=1 (119883119894119895)2 119895 = 1 2 119899 119894 = 1 2 119898 (7)

Here 119903119894119895 is the normalized preference measure of the 119894thalternative ldquo119898rdquo is the number of alternatives and ldquo119899rdquo is thenumber of criteria

International Journal of Manufacturing Engineering 5

Step 2 (construction of the weighted normalized decisionmatrix) Multiply the columns of the normalized decisionmatrix with the set of weights obtained by AHP119882 = (1199081 1199082 1199083 119908119899) to obtain weighted normalizeddecision matrix

119881 = 119877119882=(((((((

1199081 sdot 11990311 1199082 sdot 11990312 sdot sdot sdot 119908119899 sdot 11990311198991199081 sdot 11990321 1199082 sdot 11990322 sdot sdot sdot 119908119899 sdot 1199032119899 1199081 sdot 1199031198981 1199082 sdot 1199031198982 sdot sdot sdot 119908119899 sdot 119903119898119899)))))))

(8)

Step 3 (determination of the ideal and negative ideal solu-tions) The ideal solution and negative ideal solution valuesets are determined respectively as follows

119881+1 119881+2 119881+119899 = (max119894119881119894119895 | 119869 isin 119870) (min

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898 119881minus1 119881minus2 119881minus119899 = (min

119894119881119894119895 | 119869 isin 119870) (max

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898

(9)

where 119870 = 119895 = 1 2 3 119899119895 is associated with benefit criteria 1198701015840 = 119895 = 1 2 3 119899119895 is associated with cost criteria (10)

Step 4 (measurement of separation distances from idealand negative ideal solutions) Euclidean distances for eachalternative are respectively calculated as

119878+119894 = 119899sum119895=1 (119881119894119895 minus 119881+119895 )212 119894 = 1 2 119898

119878minus119894 = 119899sum119895=1 (119881119894119895 minus 119881minus119895 )212 119894 = 1 2 119898 (11)

Step 5 (calculation of the relative closeness to the idealsolution) The relative closeness to the ideal solution can bedefined as119862119894 = 119878minus119894119878+119894 + 119878minus119894 119894 = 1 2 119898 0 le 119862119894 le 1 (12)

The higher the closeness means the better the rank

Step 6 (ranking of the preference order) The preferenceorder is ranked on the basis of the order of 119862119894 Hence theoptimum alternative is the one which is nearer to the idealsolution and farther to the negative ideal solution

3 Result and Discussion

31 Computational Results The normalized decision matrixfor filler alloy electrode and shielding gas is calculated using(7) as shown in matrices (13) (14) and (15) respectively

[[[[[[[[[[[[[[[[

05351 04262 05064 05740 03994 04702 04821 0540103156 05632 04170 03553 04734 05531 05562 0465905900 04884 05064 05412 04586 05116 05191 0474804528 03918 05064 04538 05089 03318 03337 0385802470 03303 02383 02132 03846 03180 02781 02967

]]]]]]]]]]]]]]]] (13)

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Manufacturing Engineering 3

Table 3 List of attributes and alternates for shielding gas [7 8]

Properties Pure argon Pure helium 25 argon + 75helium 75 argon + 25 helium 50 argon + 50 helium

(1) Thermal conductivity 0016 01513 01174 00498 00836WmK at (1013 bar at 0∘C)(2) Gas density (lbwft) at 0∘C1 atm 01114 00111 00361 00863 00612

(3) Specific gravity (gcm3) air =l 138 01368 04476 10692 07584

(4) Ionization potential (eV) 157 245 223 179 201(5) Cost (per 100 g) ($) 05 52 4025 1675 415(6) Specific heat capacity at 0520 519 40225 16875 2855kJKgsdotK C119901 (1013 bar and 25∘C)

Table 4 Nine-point intensity of importance scale

Relative importance (119886119894119895) Description (119894 over 119895)1 Equal importance3 Moderate importance5 Strong importance7 Very strong importance9 Absolute importance2 4 6 8 Intermediate values

The multiple pairwise comparisons are based on a stan-dardized comparison scale of nine levels (Table 4)

Let 119862 = 119862119895 | 119895 = 1 2 119899 be the set of criteriaThe result of the pairwise comparison on 119899 criteria can be

summarized in an (119899times119899) evaluationmatrix119860 Every element119886119894119895 (119894 119895 = 1 2 119899) denotes the relative importance of factor119894 with respect to factor 119895 A criterion compared with itself isalways assigned the value 1 so themain diagonal entries of thepairwise comparison matrix are all 1

Square matrix119860 containing every element 119886119894119895 is shown in

119860 =( 1 11988612 sdot sdot sdot 119886111989811988621 1 sdot sdot sdot 1198862119898 sdot sdot sdot 1198861198981 1198861198982 sdot sdot sdot 1 ) 119886119894119895 = 1119886119895119894 119886119895119894 =0 (1)

(i) The relative normalized weight (119882119894) of each factor isevaluated by finding the geometric mean of 119894th rowand normalizing the geometric means of rows in thecomparison matrix

GM119894 = 1198861198941 times 1198861198942 times 1198861198943 times sdot sdot sdot times 1198861198941198951119899 119882119894 = GM119894sum119895=119899119895=1 GM119894 (2)

Matrix 119883 is obtained to denote an 119899-dimensional columnvector The sum of the weighted values is computed for theimportance degrees of alternatives then119883 = 119860 lowast119882 where119882 = [119882111988221198823 119882119899]119879 119883 = 119860 lowast119882 = [[[[[[[

1 11988612 sdot sdot sdot 119886111989911988621 1 sdot sdot sdot 1198862119899 sdot sdot sdot 1198861198991 1198861198992 sdot sdot sdot 1]]]]]]][[[[[[[11988211198822119882119899]]]]]]] =

[[[[[[[11986211198622119862119899]]]]]]]

(3)

(ii) The consistency values (CV) are calculated for thecluster of alternatives represented by the vector

CV119894 = 119862119894119882119894 (4)

(iii) The largest eigenvalue 120582max is obtained which is themean of the consistency values

(iv) The consistency index (CI) is calculated by

CI = 120582max minus 119899119899 minus 1 (5)

The AHP output quality is strictly related to theconsistency of the pairwise comparison judgements

(vi) The random index (RI) is obtained for the number offactors being used in decision making (Table 5)

(vii) The final consistency ratio (CR) is calculated as theratio of the CI and CR

CR = CIRI (6)

The acceptable upper limit for CR is 01 The evaluationprocess has to be repeated to improve the consistency if thefinal consistency ratio exceeds this value This consistencycheck is essential to establish the validity of the pairwisecomparison matrix evaluation

TOPSIS method consists of the following steps

4 International Journal of Manufacturing Engineering

Consumable material

(1) Filler alloy(2) Electrode(3) Shielding gas

Evaluation of attributes for each

material

AHP

Developing decision matrix

Evaluation and ranking of each

material

Decision of optimumconsumable material selction

Optimum filler alloy optimumelectrode optimum shielding gas

TOPSIS

Dissimilar aluminum alloy joint5083-O + 6061-T651

Applications Pressure vessels cargooil tanks gun mount base

Figure 1 Flow chart depicting integrated AHP-TOPSIS methodology for consumable materials selection

Table 5 Random index (RI) values

Criteria RI Criteria RI3 052 7 1354 089 8 1405 111 9 1456 125 10 149

Step 1 (construction of the normalized decision matrix)For the Euclidean length of a vector the element 119903119894119895 of the

normalized decisionmatrix119877 is evaluated using the followingtransformation

119903119894119895 = 119883119894119895radicsum119898119894=1 (119883119894119895)2 119895 = 1 2 119899 119894 = 1 2 119898 (7)

Here 119903119894119895 is the normalized preference measure of the 119894thalternative ldquo119898rdquo is the number of alternatives and ldquo119899rdquo is thenumber of criteria

International Journal of Manufacturing Engineering 5

Step 2 (construction of the weighted normalized decisionmatrix) Multiply the columns of the normalized decisionmatrix with the set of weights obtained by AHP119882 = (1199081 1199082 1199083 119908119899) to obtain weighted normalizeddecision matrix

119881 = 119877119882=(((((((

1199081 sdot 11990311 1199082 sdot 11990312 sdot sdot sdot 119908119899 sdot 11990311198991199081 sdot 11990321 1199082 sdot 11990322 sdot sdot sdot 119908119899 sdot 1199032119899 1199081 sdot 1199031198981 1199082 sdot 1199031198982 sdot sdot sdot 119908119899 sdot 119903119898119899)))))))

(8)

Step 3 (determination of the ideal and negative ideal solu-tions) The ideal solution and negative ideal solution valuesets are determined respectively as follows

119881+1 119881+2 119881+119899 = (max119894119881119894119895 | 119869 isin 119870) (min

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898 119881minus1 119881minus2 119881minus119899 = (min

119894119881119894119895 | 119869 isin 119870) (max

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898

(9)

where 119870 = 119895 = 1 2 3 119899119895 is associated with benefit criteria 1198701015840 = 119895 = 1 2 3 119899119895 is associated with cost criteria (10)

Step 4 (measurement of separation distances from idealand negative ideal solutions) Euclidean distances for eachalternative are respectively calculated as

119878+119894 = 119899sum119895=1 (119881119894119895 minus 119881+119895 )212 119894 = 1 2 119898

119878minus119894 = 119899sum119895=1 (119881119894119895 minus 119881minus119895 )212 119894 = 1 2 119898 (11)

Step 5 (calculation of the relative closeness to the idealsolution) The relative closeness to the ideal solution can bedefined as119862119894 = 119878minus119894119878+119894 + 119878minus119894 119894 = 1 2 119898 0 le 119862119894 le 1 (12)

The higher the closeness means the better the rank

Step 6 (ranking of the preference order) The preferenceorder is ranked on the basis of the order of 119862119894 Hence theoptimum alternative is the one which is nearer to the idealsolution and farther to the negative ideal solution

3 Result and Discussion

31 Computational Results The normalized decision matrixfor filler alloy electrode and shielding gas is calculated using(7) as shown in matrices (13) (14) and (15) respectively

[[[[[[[[[[[[[[[[

05351 04262 05064 05740 03994 04702 04821 0540103156 05632 04170 03553 04734 05531 05562 0465905900 04884 05064 05412 04586 05116 05191 0474804528 03918 05064 04538 05089 03318 03337 0385802470 03303 02383 02132 03846 03180 02781 02967

]]]]]]]]]]]]]]]] (13)

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

4 International Journal of Manufacturing Engineering

Consumable material

(1) Filler alloy(2) Electrode(3) Shielding gas

Evaluation of attributes for each

material

AHP

Developing decision matrix

Evaluation and ranking of each

material

Decision of optimumconsumable material selction

Optimum filler alloy optimumelectrode optimum shielding gas

TOPSIS

Dissimilar aluminum alloy joint5083-O + 6061-T651

Applications Pressure vessels cargooil tanks gun mount base

Figure 1 Flow chart depicting integrated AHP-TOPSIS methodology for consumable materials selection

Table 5 Random index (RI) values

Criteria RI Criteria RI3 052 7 1354 089 8 1405 111 9 1456 125 10 149

Step 1 (construction of the normalized decision matrix)For the Euclidean length of a vector the element 119903119894119895 of the

normalized decisionmatrix119877 is evaluated using the followingtransformation

119903119894119895 = 119883119894119895radicsum119898119894=1 (119883119894119895)2 119895 = 1 2 119899 119894 = 1 2 119898 (7)

Here 119903119894119895 is the normalized preference measure of the 119894thalternative ldquo119898rdquo is the number of alternatives and ldquo119899rdquo is thenumber of criteria

International Journal of Manufacturing Engineering 5

Step 2 (construction of the weighted normalized decisionmatrix) Multiply the columns of the normalized decisionmatrix with the set of weights obtained by AHP119882 = (1199081 1199082 1199083 119908119899) to obtain weighted normalizeddecision matrix

119881 = 119877119882=(((((((

1199081 sdot 11990311 1199082 sdot 11990312 sdot sdot sdot 119908119899 sdot 11990311198991199081 sdot 11990321 1199082 sdot 11990322 sdot sdot sdot 119908119899 sdot 1199032119899 1199081 sdot 1199031198981 1199082 sdot 1199031198982 sdot sdot sdot 119908119899 sdot 119903119898119899)))))))

(8)

Step 3 (determination of the ideal and negative ideal solu-tions) The ideal solution and negative ideal solution valuesets are determined respectively as follows

119881+1 119881+2 119881+119899 = (max119894119881119894119895 | 119869 isin 119870) (min

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898 119881minus1 119881minus2 119881minus119899 = (min

119894119881119894119895 | 119869 isin 119870) (max

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898

(9)

where 119870 = 119895 = 1 2 3 119899119895 is associated with benefit criteria 1198701015840 = 119895 = 1 2 3 119899119895 is associated with cost criteria (10)

Step 4 (measurement of separation distances from idealand negative ideal solutions) Euclidean distances for eachalternative are respectively calculated as

119878+119894 = 119899sum119895=1 (119881119894119895 minus 119881+119895 )212 119894 = 1 2 119898

119878minus119894 = 119899sum119895=1 (119881119894119895 minus 119881minus119895 )212 119894 = 1 2 119898 (11)

Step 5 (calculation of the relative closeness to the idealsolution) The relative closeness to the ideal solution can bedefined as119862119894 = 119878minus119894119878+119894 + 119878minus119894 119894 = 1 2 119898 0 le 119862119894 le 1 (12)

The higher the closeness means the better the rank

Step 6 (ranking of the preference order) The preferenceorder is ranked on the basis of the order of 119862119894 Hence theoptimum alternative is the one which is nearer to the idealsolution and farther to the negative ideal solution

3 Result and Discussion

31 Computational Results The normalized decision matrixfor filler alloy electrode and shielding gas is calculated using(7) as shown in matrices (13) (14) and (15) respectively

[[[[[[[[[[[[[[[[

05351 04262 05064 05740 03994 04702 04821 0540103156 05632 04170 03553 04734 05531 05562 0465905900 04884 05064 05412 04586 05116 05191 0474804528 03918 05064 04538 05089 03318 03337 0385802470 03303 02383 02132 03846 03180 02781 02967

]]]]]]]]]]]]]]]] (13)

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Manufacturing Engineering 5

Step 2 (construction of the weighted normalized decisionmatrix) Multiply the columns of the normalized decisionmatrix with the set of weights obtained by AHP119882 = (1199081 1199082 1199083 119908119899) to obtain weighted normalizeddecision matrix

119881 = 119877119882=(((((((

1199081 sdot 11990311 1199082 sdot 11990312 sdot sdot sdot 119908119899 sdot 11990311198991199081 sdot 11990321 1199082 sdot 11990322 sdot sdot sdot 119908119899 sdot 1199032119899 1199081 sdot 1199031198981 1199082 sdot 1199031198982 sdot sdot sdot 119908119899 sdot 119903119898119899)))))))

(8)

Step 3 (determination of the ideal and negative ideal solu-tions) The ideal solution and negative ideal solution valuesets are determined respectively as follows

119881+1 119881+2 119881+119899 = (max119894119881119894119895 | 119869 isin 119870) (min

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898 119881minus1 119881minus2 119881minus119899 = (min

119894119881119894119895 | 119869 isin 119870) (max

119894119881119894119895 | 119869 isin 1198701015840) | 119894= 1 2 119898

(9)

where 119870 = 119895 = 1 2 3 119899119895 is associated with benefit criteria 1198701015840 = 119895 = 1 2 3 119899119895 is associated with cost criteria (10)

Step 4 (measurement of separation distances from idealand negative ideal solutions) Euclidean distances for eachalternative are respectively calculated as

119878+119894 = 119899sum119895=1 (119881119894119895 minus 119881+119895 )212 119894 = 1 2 119898

119878minus119894 = 119899sum119895=1 (119881119894119895 minus 119881minus119895 )212 119894 = 1 2 119898 (11)

Step 5 (calculation of the relative closeness to the idealsolution) The relative closeness to the ideal solution can bedefined as119862119894 = 119878minus119894119878+119894 + 119878minus119894 119894 = 1 2 119898 0 le 119862119894 le 1 (12)

The higher the closeness means the better the rank

Step 6 (ranking of the preference order) The preferenceorder is ranked on the basis of the order of 119862119894 Hence theoptimum alternative is the one which is nearer to the idealsolution and farther to the negative ideal solution

3 Result and Discussion

31 Computational Results The normalized decision matrixfor filler alloy electrode and shielding gas is calculated using(7) as shown in matrices (13) (14) and (15) respectively

[[[[[[[[[[[[[[[[

05351 04262 05064 05740 03994 04702 04821 0540103156 05632 04170 03553 04734 05531 05562 0465905900 04884 05064 05412 04586 05116 05191 0474804528 03918 05064 04538 05089 03318 03337 0385802470 03303 02383 02132 03846 03180 02781 02967

]]]]]]]]]]]]]]]] (13)

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

6 International Journal of Manufacturing Engineering

[[[[[[[[[00021 07691 04781 07735 05927 05203 03780 0540105383 04682 03028 03970 04533 03251 06300 0465906625 02657 04143 01803 03487 03990 05040 0474805176 02277 04542 01874 03835 05203 02520 0385800621 02593 05498 04201 04184 04403 03780 02967

]]]]]]]]] (14)

[[[[[[[[[00743 07042 07041 03452 00705 0062707024 00702 00698 05387 07038 0652505450 02282 02284 04903 05455 0505002312 05455 05455 03936 02288 0210203881 03869 03869 04420 03871 05207

]]]]]]]]] (15)

These weights of all attributes of respective consumables areevaluated using AHPmethod and the results are presented inTable 6

A weighted normalized decision matrix for filler alloyelectrode and shielding gas is presented in matrices (16)

[[[[[[[[[00174 00100 00241 00608 01323 00739 01112 0042700103 00132 00199 00376 01568 00869 01283 0036800192 00115 00241 00573 01519 00804 01197 0037500148 00092 00241 00481 01686 00522 00770 0030500081 00078 00113 00226 01274 00500 00641 00234

]]]]]]]]]

[[[[[[[[[00001 00344 00755 00797 01417 01842 00255 0042700167 00209 00478 00409 01083 01151 00425 0036800206 00119 00655 00186 00833 01413 00340 0037500161 00102 00718 00193 00917 01842 00170 0030500019 00116 00869 00433 01000 01558 00255 00234

]]]]]]]]]

[[[[[[[[[00032 00708 01123 00221 00177 0024000301 00071 00111 00345 01762 0249500233 00230 00364 00314 01366 0193100099 00549 00870 00252 00573 0080400166 00389 00617 00283 00969 01991

]]]]]]]]]

(16)

32 Ranking Results The relative closeness to the idealsolution hence can be foundusing (12)The ranks are assignedbased on their ldquo119862rdquo values and are given in Table 7 The largerthe value of closeness the better the rank

It can be concluded from Table 7 filler alloy ER5356 hasthe highest ranking followed by ER5183 while ER5654 is theleast ranked material In arc electrode selection problem theoptimum arc electrode is 08 zirconiated electrode whichis the best alternate as it has slightly higher ldquo119862rdquo value than2 thoriated electrode Shielding gas analysis results indicatepure argon as the optimum consumable component closely

followed by Ar-He gas blend as second best alternative Theblock diagram for material selection application has beenshown in Figure 2

33Welding Experimentation Thebutt welding of aluminumalloy of 5083-O and 6061-T651 using pulse TIG welding (Tri-ton 220 ACDC) is performed to establish the applicabilityand validity of the selection model

The chemical composition of filler alloys used has beendepicted in Table 8

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Manufacturing Engineering 7

Table 6 Weights of the attributes from AHP method

Attributes Filler alloy weights Electrode weights Shielding gas weights1198601 00326 00311 004281198602 00235 00447 010061198603 00476 01586 015951198604 01059 01036 006401198605 03313 02399 025041198606 01572 03542 038241198607 02306 006751198608 00709

Table 7 TOPSIS analysis result

Filler alloy Electrode Shielding gasAlternates 119862119894 Rank Alternates 119862119894 Rank Alternates 119862119894 RankER5356 (C1) 08358 1 Pure tungsten (1198621) 04084 4 Pure argon (1198621) 07149 1ER5556 (C2) 06184 3 2 thoriated (1198622) 05789 2 Pure helium (1198622) 02850 4ER5183 (C3) 07205 2 2 ceriated (1198623) 04745 3 25 argon + 75 helium (1198623) 03280 3ER5654 (C4) 03142 5 15 lanthanated (1198624) 02835 5 75 argon + 25 helium (1198624) 06719 2ER5554 (C5) 03516 4 08 zirconiated (1198625) 05869 1 50 argon + 50 helium (1198625) 02363 5

Table 8 Chemical composition of filler alloys used

Composition Si Mg Fe Cu Mn Cr Zn Ti AlER5356 023 50 040 005 012 09 010 013 BalanceER5556 025 51 040 010 075 013 025 013 BalanceER5183 040 48 040 010 075 015 025 015 BalanceER5654 022 35 023 005 001 025 020 010 BalanceER5554 025 27 040 010 075 013 025 013 Balance

Experimental validation has been conducted using high-est ranking materials namely ER5356 as filler alloy 08zirconiated electrode and pure argon shielding gas as shownin Figure 3

331 Radiography Examination Welding was examinedusing radiography to check lack of penetration as shown inFigure 4 No visual defects are observed in radiography filmindicating proper fusion in weld joint leading to successfulwelding 10 times 12 cm is the scale bar used

332 Microstructure Characterizations Microstructuralcharacterization is important since various characteristics ofphase particles especially mechanical properties corrosionresistance are dependent on form of Mg2Si in which it existsin a particular material

Microstructural characterization uses optical microscopeand SEM (scanning electron microscope) images (3700NHitachi) to check grain refining and formation of brittle phaseshown in Figure 5

Microstructure shows three different regions that is basemetal heat affected zone and fusion zone on either sideof weld Figure 6 shows fusion zone (FZ) region in whichfine equiaxed grains of aluminum solid solution containing

soluble phase consisting ofMg2Si particles show proper grainrefinement Hence improved weld quality is accomplished

4 Conclusion

Selection of welding process consumables is important tasksof decision making in engineering design There is a require-ment to study the correlation between available materialalternates and their attributes to produce a robust weldingof aluminum alloys at the lowest overall cost AHP-TOPSISbased MADM model makes it possible to provide bothqualitative and quantitative study of intangible factors inwelding process

The relevance of such decision support tool lies inenabling weldingdesign engineers to improvise the processmanagement with minimum available data

In future research the approach can be used for selectionof dissimilar metal alloys for common extrusion applications

Competing Interests

The authors declare that they have no competing interests

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

8 International Journal of Manufacturing Engineering

0203040506070809

02

Clos

enes

s coe

ffici

ent

03040506070809

02

1 2 3Alternates

4 5

03040506070809

ER5356

ER5554 ER5556

ER5183

ER5654

Pure helium

08 zirconiated

2 ceriated2 thoriated

Pure tungsten15 lanthanated

Pure argon

CBA

Shielding gasElectrodeFiber alloy

75 argon + 25 helium25 argon + 75 helium

50 argon + 50 helium

Figure 2 Block diagram for material selection application

Optimum fillerER5356

Optimum arc electrode

08zirconiated

Optimum shieldinggas

Pure argon

Radiography

Best qualitydissimilar weld joint

5083-O + 6061-T651

Microstructural characterization

Optical microstructureSEM EDS

Applications

Pressure vesselscargo oil tanksgun mount bases

Figure 3 Flow diagram depicting applications of 5083-O and 6061-T651

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of Manufacturing Engineering 9

10 times 12 cm 10 times 12 cm

120 times 150 times 625mm 120 times 150 times 625mm

(a) (b)

Figure 4 (a) Weld joint 5083-O + 6061 T-651 (b) Radiography

5083-O(BM) BM BM (BM)

6061-T651

HAZ HAZFZ

5 120583m 5 120583m 5 120583m 5 120583m 5 120583m

Figure 5 Microstructure (200x) and SEM (5120583m) images of base metal heat affected zone and fusion zone at both sides

Mg2Si particles

Figure 6 SEM image of Mg2Si particle

References

[1] K Mutombo Corrosion of fatigue behaviour of 5083-H111 and6061-T651 Aluminium alloy welds [PhD thesis] University ofPretoria South Africa 2011

[2] R Scott Funderburk ldquoKey concepts in welding engineering-post weld heat treatmentrdquo Welding Innovation vol 15 no 21998

[3] R Khorshidi A Hassani A H Rauof and M Emamy ldquoSelec-tion of an optimal refinement condition to achieve maximumtensile properties of Alndash15Mg2Si composite based on TOPSISmethodrdquoMaterials amp Design vol 46 pp 442ndash450 2013

[4] S M Mirhedayatian S E Vahdat M J Jelodar and R FSaen ldquoWelding process selection for repairing nodular cast ironengine block by integrated fuzzy data envelopment analysis andTOPSIS approachesrdquoMaterials andDesign vol 43 pp 272ndash2822013

[5] C Hwang and K Yoon Multiple Attribute Decision MakingMethods and Application Survey Springer Berlin Germany1981

[6] T L Satty ldquoDecision making with the analytic hierarchyprocessrdquo International Journal of Services Sciences vol 1 pp 83ndash98 2008

[7] H-Y Huang ldquoEffects of shielding gas composition and activat-ing flux on GTAW weldmentsrdquo Materials and Design vol 30no 7 pp 2404ndash2409 2009

[8] H-Y Huang ldquoArgon-hydrogen shielding gas mixtures foractivating flux-assisted gas tungsten arc weldingrdquoMetallurgicalandMaterials Transactions A PhysicalMetallurgy andMaterialsScience vol 41 no 11 pp 2829ndash2835 2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of