optimization of process parameters for optimal mrr · pdf file234 int. j. mech. eng. &...

231

Int. J. Mech. Eng. & Rob. Res. 2014 Sujit Kumar Jha, 2014

ISSN 2278 – 0149 www.ijmerr.com

Vol. 3, No. 3, July, 2014

© 2014 IJMERR. All Rights Reserved

Research Paper

OPTIMIZATION OF PROCESS PARAMETERS FOROPTIMAL MRR DURING TURNING STEEL BAR

USING TAGUCHI METHOD AND ANOVA

Sujit Kumar Jha1*

*Corresponding Author: Sujit Kumar Jha � [email protected]

In this paper Taguchi method has been applied for optimizing process parameters namely,cutting speed, feed rates and depth of cut during turning of mild steel bar with TIN-coatedcarbide tools. The analysis of variance (ANOVA) has been applied to identify the significantprocess parameters influencing material removal rate (MRR). An orthogonal array has beenconstructed to find the experimental results of machining and further signal-to-noise (S/N) ratiohas been computed to construct the analysis of variance (ANOVA) table to study the performancecharacteristics in dry turning operations. The results of ANOVA analysis have shown that feedhas most significant factor on material removal rate (MRR) compare to depth of cut and speedfor steel. The confirmation experiments have conducted to validate the optimal parameters andimprovement of MRR from initial conditions is 347.2%.

Keywords: Taguchi method, Turning, Cutting parameters, ANOVA, MRR

INTRODUCTION

The optimization techniques have beenapplied to find out the right level or value ofthe parameters that have to be maintainedfor obtaining quality products/services. Theoptimal selection of the process parameterscan be introduced during early stage of theproduct and process development to achievethe quality product with cost effectiveness.In today's competitive and dynamic marketenvironment, manufacturing industries havefrequently assigned a high priority to

1 Engineering Department, Ibra College of Technology, Ibra, Sultanate of Oman.

economic machining under complexmachining conditions for the optimization ofproduction activities. Manufacturing the highquality product with increasing productivityis the main concern of metal-basedindustries. Productivity can be interpreted interms of MRR and in machining conditionsquality can be interpreted in terms of surfaceroughness. An increase in productivity resultsin decrease in machining time which mayresult in quality loss and vice-versa. The highspeed machining (HSM) and modernmachining technologies are being used to

232


machine the parts that need significantamount of material removal. Turning is oneof the most important machining process inwhich a single point cutting tool removesunwanted material from the surface of arotating cylindrical work piece described by(Rao, 2008).

The present study applied Taguchi methodto find out which process parameters havingmore influence on maximum materialremoval rate while turning mild steel bar workpiece on EMCO MAT 20D conventional lathe.The paper has considered three cuttingparameters namely cutting speed, feed rateand depth of cut to optimize the economicsof machining operations based onproductivity, which further depends on MRR.Higher MRR is desired by the industry forfast production in short time, which can beimproved by increasing the processparameters namely cutting speed, feed anddepth of cut. However, high cutting speedneed more power and at the same timetemperature between tool and work-pieceincrease, which give detrimental in both theproduct and tool. So, selection of suitableprocess parameters plays a vital role inefficiency and overall economy of the productto achieve the higher MRR. This paperdepicts the turning of mild steel withparameters of turning at three levels andthree factors each as considered by someresearchers like (Thamizhmanii et al., 2007;Kadirgama et al., 1996). (Haddad andFadaei, 2008) have investigated the effectsof machining parameters on material removalrate in cylindrical wire discharge turningprocess using ANOVA and found that highMRR can be obtained by fixing power and

voltage parameters as high as possible.Regardless of early works on setting upoptimum cutting speeds, feeds onconventional laths or ComputerizedNumerical Controlled (CNC) machining, therecent research (Kadirgama et al., 1996;Sanjit et al., 2010; Basim and Basir, 2010)have detailed that the process parametersneed to be optimized as CNC machining isan essential and costly process for small andmedium type manufacturing industries. Thereare many mathematical models (Yang andTarng, 1998; Taguchi et al., 2005; Taguchi etal., 1998) have been developed based byproper selection of cutting parameters andestablish the relationship between the cuttingparameters and cutting performance. Thepaper has applied Taguchi method as anotherapproach to find out the desired cuttingparameters more competently based on thedetailed description by some researchers(Abuelnaga and EI-Dardiry, 1984;Chryssolouris and Guillot, 1990). The optimalcutting parameters have determined byimplementing Taguchi method and ANOVAanalysis on S45C steel by turning operationpresented by (Yang and Tarng, 1998). Theoptimization of cutting parameters formachining hardened steel has beenobserved and found that the machining costand time can be reduced while improving thequality of the product (Gopalsamy et al.,2009).

Taguchi method is statistical methoddeveloped by Professor Genichi Taguchi forthe manufacture of robust products anddepicts the quality of a manufactured productas total loss generated by that product fromthe time it has shipped to society. The robust

233


products can be manufactured by design ofthe experiment. According to Taguchi, theprocess and product design can be improvedby identifying and setting the easilycontrollable factors. By setting those factorsat their optimal levels, the product can bemade robust to changes in operating andenvironmental conditions. Presently, Taguchimethod is applied in many sectors likeengineering, biotechnology, marketing andadvertising. According to Taguchi, there aretwo major tools namely, Signal-to-noise (S/N) ratio and orthogonal array has beenapplied in robust design. Signal to noise ratiois log functions of desired output measuresquality with emphasis on variation, andorthogonal arrays, provide a set of wellbalanced experiments to accommodatemany design factors for optimal settings ofcontrol parameters (Kaladhari and Subbaiah,2011; Park, 1989). Taguchi method has beenimplemented for optimizing machiningparameters more competently in turningprocesses by (Aman and Hari, 2005; Sujit etal., 2012). Some researchers have studiedthe influence of process parameters onperformance of various aspects of machininglike: tool life, tool wear, interaction of cuttingforces, surface roughness, material removalrate, machine tool chatter and vibration etc(Abhang and Hameedullah, 2011; Khorasaniet al., 2011; Metin, 2011). (Ezugwu andOkeke, 2010) have investigated themachining of nickel based C-263 alloy at highspeed using titanium nitride coated carbideinserts and found that feed rate is moresignificant than depth of cut in terms of toollife and its performance during machiningoperation. An experimental analysis has beendone to find out the variation of machining

parameters on MRR, gap width and surfaceroughness and graphical results has beenpresented and analyzed by Liao et al. (1997).(Lok and Lee, 1997) have evaluated themachining performance in terms of MRR andsurface roughness on ceramics using wireelectrical discharge machining. (Krishankantet al., 2012) have investigated the effects ofmachining parameters on EN24 steel byapplying Taguchi method to optimize thematerial removal rate. (Ashok et al., 2012)applied Taguchi method for optimization ofprocess parameters in turning AISI 1040 steeland found the impacts of cutting parametersnamely cutting speed, feed and depth of cutfor surface roughness and material removalrate. They found the cutting speed as themost significant cutting parameter for surfaceroughness and MRR followed by feed.(Kamal et al., 2012) did experimental analysisof turning medium Brass alloy on CNCmachine and investigated the effects ofcutting parameters on material removal rateand finally based on ANOVA analysis foundthat that feed rate has most significant effectson MRR.

Several literature reviews have beenconducted in order to meet the objectives ofthis research, to provide backgroundinformation for paper. This paper has twoobjectives to investigate the processparameters, first is to demonstrate amethodical process of using Taguchiparameter design in turning process. Thesecond is to demonstrate the use of Taguchiparameter design in order to find out theoptimum MRR with a particular combinationof cutting parameters in a turning operation.The statistical analysis techniques have beenused to assess the impacts of cutting

234


parameters on MRR. The proper selectionof process parameters is essential for gettinghigh cutting performance. The cuttingparameters are reflected on MRR, which isused to determine and to evaluate theproductivity of a turning product.

The structure of the paper is as follows.Section 2 described in details TaguchiMethod, Section 3 demonstrated Study ofmaterial and experimental setup for thisresearch followed by Data analysis resultsand discussions in Section 4. Finally, Section5 presents conclusions.

TAGUCHI METHOD

The information regarding behavior of a givenprocess can be determined by executing theexperiments on it and further collecting databased on the plan of Taguchi method. Thecollected data from all the experiments isanalyzed to study the effects of variousdesign parameters. Orthogonal arraysemployed by Taguchi method is an importanttechnique for robust design, which allow theeffects of several parameters can bedetermined efficiently with a small numberof experiments. The deviation of theexperimental value from the desired valuecan be determined by defining a loss function.The value of loss function is furthertransformed into a signal-to-noise ratio.Normally, the performance characteristic inthe analysis has been divided into threecategories, namely, the nominal-the-better,the smaller-the-better, and the higher-the-better. According to S/N analysis, the S/Nratio has been computed for each level ofprocess parameters. Despite of thecategories of the performance characteristic,

the larger the S/N ratio corresponds to thebetter performance characteristic for MRR.Further, a statistical analysis of variance(ANOVA) is performed to identify whichprocess parameters are more significant. S/N ratio is expressed on a decibel scale.Followings are the concept behind the:

• Quadratic Loss Function - used toquantify the loss incurred by the userdue to deviation from targetperformance.

• Signal-to-Noise (S/N) Ratio - used forpredicting the field quality throughlaboratory experiments.

• Orthogonal Arrays (OA) - used forgathering dependable informationabout control factors with a reducednumber of experiments.

The parametric design has been carriedout by implementing the Taguchi method todetermine an ideal feed rate and desiredforce combination, the experimental resultsshowed that surface roughness decreaseswith a slower feed rate and larger grindingforce, respectively presented by Liu et al.,(2005). The detailed description of findingoptimal cutting parameters during machiningprocesses has been presented by Taguchiet al., (1989). Taguchi has focused in the areaof quality loss functions (QLFs), orthogonalarrays (OAs), robust designs, and Signal-to-Noise (S/N) ratios. Generally, on the shopfloor technicians applied this method toimprove the quality of products andprocesses. The objective function used in theS/N ratio is maximized, by moving designtargets toward the middle of the design spacefor reducing the effects of the externalvariation (Taguchi et al., 2005).

235


Antony and Kaye (1999) haveimplemented Taguchi method by considering14 design parameters to develop new productby optimizing the manufacturing processesand improved customer satisfaction. (Zhanget al., 2007) have applied Taguchi method,which allows controlling the deviation causedby the uncontrollable factors was not includedat the conventional design of experiment. Thenumber of controllable cutting parametersduring the experiment is based on orthogonalarray (OA), which analyses the data andfinally identify the optimal condition(Abuelnaga and EI-Dardiry, 1984). (Antonyet al., 2006) have described S/N ratio usedin Taguchi method as a response of theexperiment, which is a measure of variationcaused by uncontrolled noise factors presentin the system. Noise is the result of the qualitycharacteristics influenced by external factorsduring the test. S/N ratio for the undesiredrandom noise value can be described as thesignal ratio. According to Taguchi, S/N ratiohas been used to measure the qualitycharacteristics deviating from the preferredvalue. For each level of process parametersS/N ratio has been computed and identifiedthe highest S/N ratio for the result. The higherS/N ratio value stands for the signal is muchhigher than the random effect of noise factors.There are three categories of qualitycharacteristics during analysis of the S/N ratioare given as:

• Nominal-the-Best (NB) - nearer to thetarget value is better.

• Lower-the-Better (LB) - it forecastvalues by considering defects likesurface roughness, pin holes orunwanted by-product.

• Higher-the-Better (HB) - larger thebetter characteristics consists of thedesired output as bond strength,material removal rate, employeeparticipation and customer acceptancerate.

For implementation of Taguchi method, anorthogonal array has been designed tocalculate the influence of main parametersplaced at different rows on the result andinteraction effects by doing less numberexperimental trials (Ross, 1988). The S/Nratio measured the performancecharacteristics of the levels of control factorsagainst these factors. The category higher-the-better is used to calculate S/N ratio formaterial removal rate mention in equation (1):

...(1)

Where n is the number of values at eachtrial and yi is the each observed value.

In addition, a statistical Analysis ofVariance (ANOVA) has been used todetermine the significance of the cuttingparameters as well as to examine whichparameters are significantly affecting theresponses. The implementation ofconventional experimental design methodsare very difficult due to its complicacy andrequired more number of experiments byincreasing the process parameters describedby Kaladhari and Subbaiah (2011). Taguchimethod minimizes the number ofexperimental trials, by implementing aparticular design of orthogonal arrays to studythe entire parameter space with small numberof experiments.

10 2

1

1 1/ 10log

n

i i

S Nn y=

⎛ ⎞= − ⎜ ⎟

⎝ ⎠∑

236


Steps Involved in Taguchi Method

For larger-the-better characteristics, Taguchimethod have been used for a parameterdesign includes the following steps (Nian etal., 1999):

1. Select a suitable output qualitycharacteristic to be optimized.

2. Select the control factors and theirlevels, identify their possibleinteractions.

3. Select noise factors and their levels.

4. Select sufficient inner and outer arrays.Control factors assigned to inner arrayand noise factors to the outer array.

5. Carry out the experiment.

6. Execute statistical analysis based on S/N ratio.

7. Predict optimal output performancelevel based on optimal control factorlevel combination, and conduct aconfirmation experiment to verify theresult.

EQUIPMENT AND MATERIALS

This paper has considered work piecematerial as mild steel bar of total length 100mm and straight turning has been done on50 mm length of the work piece diameter 24mm. Young’s Modulus of mild steel is 215GPa, Poisson’s ratio is 0.29 and melting pointis 1410°C. For proper selection of cuttingparameters, a mathematical model for TIN-coated carbide tools has been developedbased on statistical techniques to establishthe relationship between the cuttingperformance and the cutting parameterspresented by Chua et al., (2006). It has beenalso studied that the effect of the length and

The experiments were conducted onEMCO MAT 20D conventional lathe, whichis highly versatile and up to date with thelatest technology. EMCO MAT 20D lathe hasa high tech 3 axes digital display, 999 toolpositions with stepless 4 staged gearbox -power 5.3 kW, 40 - 3000 rpm and simul-taneous complex axes cutting work to takeplace at the same time. The experimentswere conducted on standardized shown be-low Figure 2 the configuration of the machineas listed.

Industrial design: Swing diameter overbed: 400 mm, Swing diameter over crossslide: 250 mm, Distance between Centres:1000 mm, Spindle speed: 40 - 3000 rpm,

diameter of work piece during machiningoperations, while keeping the cutting speedconstant demonstrated by (Mustafa and Ali,2011). The material removal rate (MRR) playsan important characteristic in turningoperation and according to (Gaitonde et al.,2003; Paulo Davim, 2003) high MRR isalways desirable for increasing theproductivity. The details of experiments,experimental variables and constants havebeen shown in the Table 1.

Table 1: Experimental Details

Details of the Experiment

Machine used: EMCO MAT - 20D(Conventional Lathe)

Work-piece material: Mild SteelDensity of material: 7850 kg/m3Diameter of Work Piece: 24 mmHardness of Material: 29.8 HRC

Tool material: M42 Series HighSpeed SteelShape of Cutting Tool- Triangle

Process parameters: Speed, Feedand Depth of cut

Cutting Condition: Dry Machining

Experimentalat Variables

SpeedFeedDepth of Cut

ExperimentalConstants

Work Piece:Mild SteelCuttingconditionCNCmachineCutting Toolmaterial

237


Figure 1: EMCO MAT 20D Conventional Lathe

Spindle bore: 50 mm, Chuck diameter max:200 mm.

Selection of Cutting Parameters

and Their Levels

In this paper, there are three cuttingparameters: cutting speed, feed rate anddepth of cut are considered for three levelsas the control factors. Three variables arestudied for three levels and hence nineexperiments were designed and conductedbased on Taguchi’s L9 orthogonal array. Theinitial cutting parameters have been taken ascutting speed of 45 m/min, a feed rate of 0.11mm/rev, and a depth of cut 0.5 mm. Thecutting parameters and their levels of thisexperiment are shown in Table 2.

Table 2: Cutting Parameters and Their Levels

Processparameters

Cutting speed

Feed rate

Depth of cut

Symbol

A

B

C

Unit

m/min

mm/rev

mm

Level 1

30

0.05

0.5

Level 2

45

0.11

1.0

Level 3

60

0.22

1.5

Levels of factors

DATA ANALYSIS, RESULTS

AND DISCUSSIONS

Experimental results and their correspondingS/N ratios have been computed for each levelof process parameters and shown in Table4. ANOVA has been carried out to observethe significance of factors on responses.Lastly, confirmation test has conducted toverify the optimal results. An orthogonal arrayhas been used to find out the optimal cuttingparameters in reduce number of cuttingexperiments. The results of an orthogonalarray have been further studied by using S/N ratio and ANOVA analyses. The optimalcutting parameters have been determinedfrom the results of S/N ratio and ANOVAanalyses.

Orthogonal Array Experiment

A suitable orthogonal array has beendesigned for experiments based on thedegrees of freedom. The experimental planhas been designed based on Taguchi L9

orthogonal array as given in Table 3.

Table 3: Experimental Layout Usingan L9 Orthogonal Array

ExperimentalNo.

1

2

3

4

5

6

7

8

9

Cutting Speed

1

1

1

2

2

2

3

3

3

Feed Rate

1

2

3

1

2

3

1

2

3

Depth of cut

1

2

3

3

1

2

2

3

1

Machining parameter level

238


...(2)

Where, Wi is the initial weight of the workpiece in grams, Wf is the final weight of thework piece in grams, ρ is the density of thematerial in grams / mm3, and t is the timetaken for machining. In practice MRR shouldbe high, thus Taguchi method refers to selectthe process parameter having more S/N ratio.

The response table, which contains themean of the S/N ratios for each level and foreach factor is shown in the Table 5. For

Analysis of the Signal-to-Noise

(S/N) Ratio

In Taguchi method, S/N ratio has been usedto measure the quality of characteristicdeviating from the desired value. Asmentioned earlier, there are three categoriesof performance characteristic, namely, lower-the-better, higher-the-better, and nominal-the-better. To get optimal machining perfor-mance, the higher-the-better performance

characteristic for MRR has been taken. Table4 shows the experimental results for MRRand corresponding S/N ratio using Eq. (1) andthen separate out the effect of each cuttingparameter at different levels.

Material removal rate (MRR) can becalculated using the difference of weight ofwork piece before and after the machiningoperation.

Table 4: Experimental Results for MRR and S/N Ratio

Exp.No.

Cutting Speed

(m/min)Feed

(mm/rev)Depth of Cut

(mm)

Cycle Time toRemove

Material (sec)MRR (gm/min)

MRR (mm3min)

S/N Ratio(dB)

The total mean S/N ratio = 65.18 dB

Control Parameter (Level) Result / Observed Value

1

2

3

4

5

6

7

8

9

30

30

30

45

45

45

60

60

60

0.05

0.11

0.22

0.05

0.11

0.22

0.05

0.11

0.22

0.5

1.0

1.5

1.0

1.5

0.5

1.5

0.5

1.0

131

66

33

87

44

22

66

33

16

3.346

12.884

36.552

8.711

24.433

15.335

14.495

9.213

35.374

426.27

1641.34

4656.31

1109.68

3112.06

1953.50

1846.55

1173.74

4506.25

52.59

64.30

73.36

60.90

69.86

65.82

65.33

61.39

73.07

example, the sum of S/N ratio for cuttingspeed at levels 1, 2, and 3 has beencalculated by adding the S/N ratio for theexperiments 1-3, 4-6, and 7-9. The sum ofS/N ratio for each level of the other cuttingparameters has been computed in similarmanner. Then the average S/N ratio forcutting speed at levels 1, 2, and 3 can becomputed by averaging the S/N ratios for theexperiments 1-3, 4-6, and 7-9, which isdenoted by MS1, MS2, and MS3. MS1 is givenby:

.i fW W

MRRtρ

−=

239


Similarly the MS2 and MS3 are calculated andare shown in the table below table. MF1, MF2,

1

1 1/3MS S Nratio forthe first level speed= =∑

(52.59+64.30+73.36)=63.4213

MF3 and MD1, MD2, MD3 are the average S/N ratio for feed and depth of cut factor, forthree levels 1, 2 and 3. The mean S/N ratiofor each level of cutting parameters iscalculate and presented in Table 6, called themean S/N response table for MRR.

Table 5: S/N Response Table to Identify Significance of Machining Parameter for MRR

Machining Parameters Level 1Symbol Level 2 Level 3

3.18

11.15

9.59

Mean S/N Ratio (dB) Significance ofMachiningParameterMax-Min

66.60 *

70.75 *

69.52 *

65.52

65.18

66.09

63.42

59.60

59.93

A

B

C

Cutting speed (m/min)

Feed rate (mm/rev)

Depth of cut (mm)

Figure 2 shows the main effect on materialremoval rate which is primarily due to feedand depth of cut. The cutting speed is foundto be least significant from the main effectplot. From Table 5 and Figure 2, all level totalsare compared and combination yielding thehighest combined S/N ratio is selected formaximum metal removal rate. In thisexperiment, S3-F3-D3 combination yields themaximum metal removal rate. This is theoptimal levels combination of factors forturning operation in lathe for mild steelmaterial. Therefore, based on Table 5 andFigure 2, the optimal process parameters forMRR are as follows: Cutting speed at level 3is 60 m/min, feed at level 3 is 0.22 and depthof cut at level 3 is 1.5 mm i.e. A3-B3-C3. Theconfirmation test has been conducted withabove optimal combination of processparameters for MRR and improvement in S/N ratio as 10.80 dB has been noticed in Table7. Based on the results of confirmation test,the improvement of MRR from initial cuttingparameters is around 347.2%.

Analysis of Variance (ANOVA) For MRR

The purpose of the ANOVA is to investigatewhich process parameter has more impactson performance characteristic. F-test can be

Figure 2: Main Effect Plot for S/N Ratio (MRR)

performed by computing the mean of squareddeviations SSD and then calculating mean ofsquared deviation (MS) by dividing SSD/degrees of freedom associated with theprocess parameter. Then F-value for each

240


process parameter can be calculated as SSD/SSE. Normally, larger the F-value more effectson performance characteristic due. Al last,percentage contribution of individualparameters can be determined using ANOVAanalysis. Table 6 shows the results of ANOVAanalysis for MRR. The F-value and % C forfeed rate is more in the table, which indicatesthat feed rate is significantly contributingtowards machining performance than cuttingspeed and depth of cut.

Table 6: Results of the Analysisof Variance for MRR

Factor

Speed

Feed

Depth of Cut

Error

Total

SSD

15.69

186.48

141.67

0.90

344.75

DF

2

2

2

2

8

MS

7.845

93.24

70.835

0.45

F

17.43

207.20

157.41

%C

4.56

54.23

41.20

Therefore, based on S/N ratio and ANOVAanalysis for MRR, the optimal cuttingparameters for material removal rate hasbeen found from Table 2 at level 3 as cuttingspeed 60 m/min in level 3, feed 0.22 mm/revin level 3 and depth of cut 1.5 mm in level 3.From above analysis, it is revealed that thefactor feed rate has the most significant factorand its contribution to material removal rateis more. The next significant factor is cuttingspeed and least significant factor is depth ofcut.

Confirmation Tests

After selecting the optimum level of processparameters, the final step is to predict andverify the improvement in performancecharacteristic by using the optimum levelcutting parameters. The estimated S/N ratio

η predicted using the optimal level of the processparameters can be computed as (Yang andTarng, 1998):

...(3)

where ηm is the total mean of the S/N ratio,ηi is the mean S/N ratio at the optimal level,and o is the number of main processparameters that affect the performancecharacteristic. The estimated S/N ratio usingthe optimal cutting parameters for MRR hasbeen obtained and the corresponding MRRcan be calculated by using Eq. (2). Themeasuring data and the actual S/N ratio ofconfirmation experiments are listed in Table7. Table 7 shows the comparison betweenpredicted MRR with the actual MRR usingoptimal cutting parameter. The increase ofthe S/N ratio from the initial cutting parameterto the optimal cutting parameter is 10.8 dB,which means that MRR is increased by about3.47 times. Therefore, prior design andanalysis for optimizing the cutting parametersrequired for optimum MRR.

( )1

o

predicted m i mi

η η η η=

= + −∑

Table 7: Results of Conformation Test for MRR

Level

MRR(mm3/min)

S/N ratio (dB)

InitialCutting

Parameter Predicted

A3B3C3

6760

76.6

Experiment

A3B3C3

8076

78.14

Optimal CuttingParameter

A2B2C1

2326

67.34

Improvement of S/N ratio = 10.80 dB

CONCLUSION

The experimental results based on S/N ratioand ANOVA analysis provides a systematicand efficient methodology for the optimizationof cutting parameters for MRR. The materialremoval rate is mainly affected by cutting

241


speed, depth of cut and feed rate, byincreasing any one the material removal rateis increased. The result based on F-testshows that feed rate has more impact onperformance characteristic than cutting anddepth of cut. The best parameters for materialremoval rate have been found from Table 2at level 3 as cutting speed 60 m/min in level3, feed 0.22 mm/rev in level 3 and depth ofcut 1.5 mm in level 3. The confirmationexperiments have been conducted to validatethe optimal cutting parameters. Theimprovement of MRR from initial cuttingparameters to the optimal cutting parametersis around 347.2%.

REFERENCES

1. Abhang L B and Hameedullah M (2011),“Modeling and Analysis for SurfaceRoughness in Machining EN-31 Steelusing Response Surface Methodology”,International Journal of Applied Researchin Mechanical Engineering, Vol. 1, No. 1,pp. 33-38.

2. Abuelnaga A M and EI-Dardiry M A(1984), “Optimization Methods for MetalCutting”, International Journal of MachineTool Design and Research, Vol. 24, No.1, pp. 11-18.

3. Aman Aggarwal and Hari Singh (2005),“Optimization of Machining Techniques -A Retrospective and Literature Review”,Sadhana, Vol. 30, No. 6, pp. 699-711.

4. Antony J and Kaye M (1999),“Experimental Quality - A StrategicApproach to Achieve and ImproveQuality”, Kluwer Academic Publishers,Dordrecht.

5. Antony J, Perry D, Wang C and Kumar

M (2006), “An Application of TaguchiMethod of Experimental Design for NewProduct Design and DevelopmentProcess”, Assembly Automation, Vol. 26,No. 10, pp. 18-24.

6. Ashok Kumar Sahoo, Achyuta NandaBaral, Arun Kumar Rout and Routra B C(2012), “Multi-objective Optimization andPredictive Modeling of SurfaceRoughness and Material Removal Ratein Turning using Grey Relational andRegression Analysis”, ProcediaEngineering, Vol. 38, pp. 1606-1627.

7. Basim Khidhir A and Bashir Mohamed(2010), “Study of Cutting Speed onSurface Roughness and Chip Formationwhen Machining Nickel-based Alloy”,Journal of Mechanical Science andTechnology, Vol. 24, No. 5, pp. 1053-1059.

8. Chryssolouris G and Guillot M (1990), “AComparison of Statistical and AIApproaches to the Selection of ProcessParameters in Intelligent Machining”,ASME-Journal of Engineering forIndustry, Vol. 112, No. 2, pp. 122-131.

9. Chua M S, Rahman M, Wong Y S andLoh H T (2006), “Determination ofOptimal Cutting Conditions using Designof Experiments and OptimizationTechniques”, Int. J. Mach. Tools Manuf.,Vol. 33, No. 2, pp. 297-305.

10. Ezugwu E O and Okeke C I (2000),“Performance of PVD Coated CarbideInserts when Machining a Nimonic C-263Alloy at High Speed Conditions”,Tribology Transaction, Vol. 43, No. 2, pp.332-336.

242


11. Gaitonde V N, Karnik and Paulo Davim(2003), “Multi-performance Optimizationin Turning of Free-machining Steel usingTaguchi Method and Utility Concept”,Journal of Materials Engineering andPerformance, Vol. 18, No. 3, pp. 231.

12. Gopalsamy B M, Mondal B and Ghosh S(2009), “Taguchi Method and ANOVA: AnApproach for Process ParametersOptimization of Hard Machining WhileMachining Hardened Steel”, Journal ofScientific and Industrial Research, Vol.68, pp. 686-695.

13. Haddad M J and Fadaei Tehrani A (2008),“Material Removal Rate (MRR) Study inthe Cylindrical Wire Electrical DischargeTurning (CWEDT) Process”, Journal ofMaterials Processing Technology, Vol.199, pp. 369-378.

14. Kadirgama M M, Noor N M, Zuki M M,Rahman M R M, Rejab R, Daud K andPark S H (1996), “Robust Design andAnalysis for Quality Engineering”,Chapman & Hall, London.

15. Kaladhari M and Subbaiah K V (2011),“Application of Taguchi Approach andUtility Concept in Solving the Multi-objective Problem when Turning AISI 202Austenitic Stainless Steel”, Journal ofEngineering Science and TechnologyReview, Vol. 4, No. 1, pp. 55-61.

16. Kamal Hassan, Anish Kumar and GargM P (2012), “Experimental Investigationof Material Removal Rate in CNC Turningusing Taguchi Method”, InternationalJournal of Engineering Research andApplications, Vol. 2, No. 2, pp. 1581-1590.

17. Khorasani A M, Yazdi M R S andSafizadeh M S (2011), “Tool LifePrediction in Face Milling Machining of7075 Al by Using Artificial NeuralNetworks (ANN) and Taguchi Design ofExperiment (DOE)”, International Journalof Engineering and Technology, Vol. 3,No. 1, pp. 30-35.

18. Krishankant, Jatin Taneja, Mohit Bectorand Rajesh Kumar (2012), “Applicationof Taguchi Method for Optimizing TurningProcess by the Effects of MachiningParameters”, International Journal ofEngineering and Advanced Technology,Vol. 2, No. 1.

19. Liao Y S, Huang J T and Su H C (1997),“A Study on the Machining-parametersOptimization of Wire Electrical DischargeMachining”, Journal of MaterialsProcessing Technology, Vol. 71, No. 3,pp. 487-493.

20. Liu C H, Andrian C, Chen C A and WangY T (2005), “Grinding Force Control inAutomatic Surface Finish System”,Journal of Materials ProcessingTechnology, Vol. 170, pp. 367-373.

21. Lok Y K and Lee T C (xxxx), “Processingof Advanced Ceramics using the Wire-cut EDM Process”, Journal of MaterialsProcessing Technology, Vol. 63, No. 1-3,pp. 839-843.

22. Metin Kok (2011), “Modeling andAssessment of Some Factors thatInfluence Surface Roughness for theMachining of Particle Reinforced MetalMatrix Composites”, Arabian Journal forScience and Engineering, Vol. 36, No. 7,pp. 1347-1365.

243


23. Mustafa A Y and Ali T (2011),“Determination and Optimization of theEffect of Cutting Parameters and WorkPiece Length on the GeometricTolerances and Surface Roughness inTurning Operation”, International Journalof the Physical Sciences, Vol. 6, No. 5,pp. 1074-1084.

24. Nian C Y, Yang W H and Tarang Y S(1999), “Optimization of TurningOperations with Multiple PerformanceCharacteristics”, Journal of MaterialProcess Technology, Vol. 95, pp. 90-96.

25. Park S H (1989), “Robust Design andAnalysis for Quality Engineering”,Chapman & Hall, London.

26. Paulo Davim J (2003), “Design ofOptimization of Cutting Parameters forTurning Metal Matrix Composites Basedon the Orthogonal Arrays”, Journal ofMaterial processing Technology, Vol. 132,pp. 340-344.

27. Rao P N (2008), “ManufacturingTechnology, Tata McGraw HillPublication”, India.

28. Ross P J (1988), “Taguchi Techniques forQuality Engineering”, McGraw HillPublishers, New York.

29. Sanjit Moshat, Saurav Data, AshishBandopaddhayay and Pradip Kumar Pal(2010), “Optimization of CNC MillingProcess Parameters using PCA BasedTaguchi Method”, International Journal ofEngineering Science and Technology,Vol. 2, No. 1, pp. 92-102.

30. Sujit Kumar Jha, Sujit Singh and ShivRanjan Kumar (2012), “ExperimentalAnalysis of CNC Turning using TaguchiMethod”, Proceedings of ThirdInternational Conference on Advances inMechanical Engineering, Delhi, India,DOI: 02.AETAME, Vol. 3, No. 15, pp. 91-95.

31. Taguchi G, Choudhary S, and Wu Y(2005), “Taguchi’s Quality EngineeringHandbook”, John Wiley & Sons Inc., NewJersey.

32. Taguchi G, EI Sayed M and Hsaing C(1989), “Quality Engineering andProduction Systems”, New York:McGraw-Hill.

33. Thamizhmanii S, Saparudin S and HasanS (2007), “Analysis of SurfaceRoughness by Turning Process usingTaguchi”, Journal of Achievements inMaterials and ManufacturingEngineering, Vol. 20, No. 1-2, pp. 503-506.

34. Yang W H and Tarng Y S (1998), “DesignOptimization of Cutting Parameters forTurning Operations Based on the TaguchiMethod”, Journal of Materials ProcessingTechnology, Vol. 84, pp. 122-129.

35. Zhang J Z, Chen J C and Kirby E D(2007), “Surface RoughnessOptimization in an End-Milling Operationusing the Taguchi Design Method”,Journal of Material ProcessingTechnology, Vol. 184, pp. 233-239.

optimization of process parameters for optimal mrr · pdf file234 int. j. mech. eng. &...

Documents