ISSN (Print) : 0974-6846

Indian Journal of Science and Technology, Vol 8(36), DOI: 10.17485/ijst/2015/v8i36/82823, December 2015 ISSN (Online) : 0974-5645

Effect of Machining Parameters and Optimization of

Machining Time in Facing Operation using Response
Surface Methodology and Genetic Algorithm
R. Babu1*, D. S. Robinson Smart1, G. Mahesh2 and M. Shanmugam3
1
Department of Mechanical Engineering, Karunya School of Mechanical sciences, Karunya University, Coimbatore,
Tamil Nadu, India; [email protected]
2
Department of Mechanical Engineering, Sree Sakthi Engineering College, Karamadai, Coimbatore,
Tamil Nadu, India
3
Bimetal Bearings Limited, Coimbatore, Tamil Nadu, India

Abstract
The objective of this research is to select the machining parameters and its cutting conditions to increase the productivity
and minimize of total machining time and machining cost. A significant improvement in process may lead to increase in
the process efficiency and low cost of manufacturing. In this research, Spindle speed, Feed rate, Depth of cut and End relief
angle are considered as input parameters for facing operation of the A22E Bimetal bearing material using M42 HSS tool
material. A second order mathematical model was developed by using Design of Experiments (DoE) of Response Surface
Methodology (RSM) to predict the machining time of Bimetal bearing material. The Analysis of Variance (ANOVA) was used
to study the performance characteristics in facing operation. The direct and interaction effects of the machining parameters
were also analyzed using Design Expert software. The values of Prob > F less than 0.05 indicate model terms are significant.
The Genetic Algorithm (GA) was trained and tested by using MATLAB 7.0. The GA recommends 1.169 seconds as the best
minimum predicted machining time value. The confirmatory test shows the predicted values and experimental values
were very close and good agreement.

Keywords: Depth of Cut, End Relief Angle, Feed Rate, Genetic Algorithm (GA), Machining Time, Response Surface
Methodology (RSM), Spindle Speed

1. Introduction of tool, nose radius etc.) torque, spindle motor current,

cutting time, clearance angle, feed drive current, type of
During the machining of hardened bearing steel, the ­lubricants used etc. Optimized combination of the above
work piece and tool is subjected to high strain, tem- mentioned parameters will increase in the surface finish,
perature and vibration results low surface roughness, reduces the tool wear, machining time and increase the
tool wear, high increase in machining time etc. The tool life1. It is necessary to develop a technique to predict
machining parameters selection plays a major role in the surface roughness, tool wear, machining time etc.
production based industries. Improper selection of Nowadays in order to obtain the exact end result several
machining parameters may leads to increase in machin- modeling procedure and techniques are used, they are
ing time and production cost. To achieve the desired classified as 1) Analytical based models, 2) Experimental
results the cutting parameters such as feed rate, spindle based models, 3) DoE (Design of Experiments) based
speed, axial/radial depth of cut, tool geometry (cut- models, 4) AI (Artificial Intelligence) based models
ter diameter, number of teeth, side cutting edge angle, such as Response Surface Methodology (RSM), Genetic
rack angle, shank diameter, helix angle, overall length Algorithm (GA), Artificial Neural Network (ANN) etc.,

*Author for correspondence

 Effect of Machining Parameters and Optimization of Machining Time in Facing Operation using Response Surface Methodology
and Genetic Algorithm

were used in engineering applications particularly in to predict five-axis machining time. Machining time
machining related work2. ­calculation algorithms have been developed to assess the
This paper presents the optimization and predic- selection of tools using machining performance related
tion of machining parameters of machining on A22E criteria10. The authors proposed a method for machin-
Bimetal bearing considering the input parameter ing time prediction using a mechanistic approach11. A
such as spindle speed, feed rate, depth of cut and end mechanistic approach was proposed to estimate real
relief angle. A second order mathematical model was machining time more accurately12. An investigation was
developed using Analysis of Variance (ANOVA). The carried by the author13 to estimate the actual machin-
Response Surface Methodology (RSM) was used to ing time varies depending on the machining conditions
predict machining time and the interaction effect was in a practical operation. The authors14 investigated that
analyzed using RSM. The non-traditional optimization for machining, it is difficult to select the cutting speed,
technique Genetic Algorithm (GA) was employed for depth of cut and feed rate combination and suggests that
the optimum cutting parameters which minimizes the required depth in one pass to keep machining time and
machining time. low cost.
The mathematical models have to develop to corre- From the literature sources, there is a limited research
late the cutting parameters with cutting performance to available in hard turning and facing operation of bear-
determine the optimal cutting parameters. But yet well ing materials in order to reduce machining time. It is
known the reliable mathematical models are not easy to found that the machining of A22E (BIMETAL BEARING
obtain3. The authors4 have optimized the cutting parame- MATERIAL) metal matrix composite is an important
ters (cutting speed, feed rate and depth of cut) considering area of research.
maximum production rate and minimum production Bimetal bearings are used to support the crank shaft
cost. The authors5 expressed the equation for machining and connecting rods. Bimetal bearing are constructed of
time. two layers having a steel back, which supports the bear-
ing structure shown in Figure 1. The second layer is the
pDL bearing lining. It is relatively thick. Its thickness is about
tm =  (1)
10FV 0.012″. Commonly, the lining is made of an aluminum
Where, alloy containing 6–20% of tin. Tin serves as a solid
lubricant and provides anti-friction properties. Another
tm - Machining time per piece (min/ pc),
additive is 2–4% of silicon dispersed in aluminum in form
D - Diameter of the work piece (mm),
of fine particles. Hard silicon strengthens the alloy and
L - Length of the work piece (mm),
also serves as an abrasive polishing the journal surface.
V - Cutting speed (m/min),
Presence of silicon is particularly important for engines
F - Feed rate (mm/rev)
with cast iron crankshafts. The alloy may be addition-
and also suggests that machining time decreases with an ally strengthened by copper, nickel and other elements.
increase in cutting speed. An experimental investigation The two main layers (steel and aluminium) are bonded to
was done using aluminium to study the behavior and each other by means of a bonding layer.
effect of machining parameters such as cutting speed,
feed rate, and depth of cut on the surface roughness in
minimum machining time on face milling process6. The
authors7 conducted the experiment to establish the cor-
relation between cutting velocity, feed rate and cutting
time by using Taguchi design of experiment. The mathe-
matical model was developed by the authors8 to estimate
the machining time by measuring acceleration rate and
tool path geometry, which results in good accuracy at
high feed rate. The authors9 developed a machining time
calculation algorithm by considering feed angle and
machining speed based on machine behavior in order Figure
Figure 1. 1.Aluminium
Aluminiumalloy
alloy bearing
bearing g (Bimetal
(Bimetal beearing).
bearing).

2 Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology
 R. Babu, D. S. Robinson Smart, G. Mahesh and M. Shanmugam

2. Response Surface Methodology 3.2 Crossover

(RSM) This operator randomly chooses a locus and exchanges
the sub sequences before and after that locus between two
Response Surface Methodology (RSM) consists of a chromosomes to create two offspring. For example, the
group of mathematical and statistical techniques used in strings 10000100 and 11111111 could be crossed over
the development of an adequate functional relationship after the third locus in each to produce the two offspring
between a response of interest y, and a number of associ- 10011111 and 11100100. The crossover operator roughly
ated control (or input) variables denoted by x1, x2, . . . , mimics biological recombination between two ­single-
xk,. In general, such a relationship is unknown but can be chromosome (haploid) organisms.
approximated by a low-degree polynomial model of the
form15
3.3 Mutation
y = f ′(x) β + e (2) This operator randomly flips some of the bits in a
Where. ­chromosome. For example, the string 00000100 might
be mutated in its second position to yield 01000100.
x = (x1, x2, ….. , xk).
Mutation can occur at each bit position in a string with
f(x) is a vector function of p elements
some probability, usually very small (e.g., 0.001).
e is a random experimental error assumed to have a
The most important components in a GA consist of:
zero mean.
f ′(x) β = the mean response • Representation (definition, of individuals)
The most frequently used second-order designs are the • Evaluation function (or fitness function)
3K factorial, central composite, and the Box-Behnken • Population
designs. The steps concerned with RSM are as follows • Parent selection mechanism
• Variation operators (crossover and mutation)
1. Design the set of experiment for adequate and reliable
• Survivor selection mechanism (replacement)
measurement.
2. Define the mathematical model
3. Setting maximum and minimum response value for 4. Identification of Important
experimental factors Process Parameters
4. Direct and interaction effect of process variable of the
machining parameters. The ranges of machining parameters, conditions and
tool end relief angle were selected from the tool manu-
facturer and machining data handbook19. The author20
3. Optimization by using Genetic discussed the steps and procedure to conduct the
Algorithm (GA) experiment. The authors21 discussed for hard and tough
material, the relief angle should be 6 to 8 degrees for
Genetic Algorithm (GA) is a global optimization
HSS tools and 5 to 7 degrees for carbide tools. For
­algorithm derived from evolution and natural selection.
medium steels, mild steels, cast iron, the relief angle
Genetic algorithms tend to thrive in an environment in
should be 8 to 12 degrees for HSS tools and 5 to 10
which there is a very large set of candidate solutions and
degrees for carbide tools. For ductile materials such as
in which the search space is uneven and has many hills
copper, brass, bronze and aluminium, ferritic malleable
and valleys. GA is one of the most powerful methods with
iron, the relief angle should be 12 to 16 degrees for HSS
which to (relatively) quickly create high quality solutions
tools and 5 to 14 degrees for carbide tools. The authors
to a problem16–18.
finally concluded that larger relief angle generally tend
to produce a better surface finish. Optimal performance
3.1 Selection of any machining process is based on choosing the right
This operator selects chromosomes in the population for combination of input parameters. The independently
reproduction. The fitter the chromosome, the more times controllable process parameters affecting the machining
it is likely to be selected to reproduce. time were identified to carry the experimental work and

Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology 3
 Effect of Machining Parameters and Optimization of Machining Time in Facing Operation using Response Surface Methodology
and Genetic Algorithm

mathematical model were developed. The ­important The M42 HSS single point cutting tool is used for facing
controllable process parameters considered for this operation. The properties of M42 HSS tools are
investigation are spindle speed (rpm), feed rate (mm/
min), depth of cut (mm) and end relief angle (degree) as 1.08% -carbon,
shown in Table 1. 3.8% - Chromium,
9.4% - Molybdenum,
1.5% -Tungsten,
5. Development of Design Matrix 8% - Cobalt,
1.2% - Vanadium
The Design Of Experiment (DOE) is used to develop
a design matrix. The central composite second order The special type industrial CNC lathe was used to conduct
rotatable design was utilized to design using Design the experiments under dry condition. The machining
Expert V 9.0.4 software. The design matrix, consists experimental set-up of Bimetal Bearing experimental is
of three-level, four factor central composite rotatable shown in Figure 2.
factorial design (CCD) consisting of 30 sets of coded The response parameters machining time was mea-
conditions. The upper limit of a given parameter was sured. Table 2 shows the experimental design matrix and
coded as (+2) and the lower limit was coded as (–2). the measured response of machining time.
The intermediate levels of –1, 0, +1 of all the variables
have been calculated by interpolation. Thus, all the 30
experimental runs to allow the estimation of the linear, 7. Response Surface Model for
quadratic and two way interactive effects of the process the Prediction of Machining
parameters.
Time
RS model was developed to predict the machining time.
6. Experimental Set Up and The Design Expert V 9.0.1 software of state ease was car-
Conditions ried out with the experimental and analysis purpose.
Bimetal Bearing Specimen of size 95 mm diameters The model is checked for its adequacy using ANOVA
and thickness 3mm are selected for experimental pur- (analysis of variance). ANOVA table for the prediction of
pose. The bimetal bearing consists of steel alloy on inner Machining Time is shown in Table 3. The model is sig-
side and aluminium alloy on outerside. The aluminium nificant and the lack of fit is not significant which infers
­bonding layer is used for binding of the alloys. the significance of the model. Values of Prob> F less than
Figure 1. the
0.05 indicate A modelalloy
Aluminium a terms
bearing
asgin(Bimetal beearing).
significant and the
Bimetal bearing is softest and it consists of
values greater than 0.10 indicate the model terms as not
6 - 20% tin,
1 % copper,
2 - 4% silicon and highly strengthened by nickel and
other elements.

Table 1. Process factors and their levels

Variables Unit Coded Variable Level
Lowest Low Centre High Highest
–2 –1 0 +1 +2
Spindle Speed rpm 400 500 600 700 800
Feed Rate mm/rev 0.04 0.06 0.08 0.1 0.12
Depth of Cut mm 1 1.2 1.4 1.6 1.8
End Relief Degree 8 10 12 14 16
Figuure 2. Machiining of bim
metal bearing.
Angle
Figure 2. Machining of bimetal bearing.

4 Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology
 R. Babu, D. S. Robinson Smart, G. Mahesh and M. Shanmugam

Table 2. Experimental design matrix and output obtains from the Design Expert software in terms of
response factors actual factors are given:
Run Spindle Feed Depth of End Relief Output Machining Time (T) - + 1.13315- 4.63660E-003*
Speed (A) Rate (B) Cut Angle (D) Response A+7.18586* B-0.29985* C+0.26515* D-9.53125E-003* A *
rpm mm/rev (C) mm degree Machining
B - 6.40625E-004* A * C +4.06250E-005* A* D +1.71875*
Time (T) Sec
B *C - 0.24219* B * D+7.03125E-003* C * D + 4.49781E-
1 500 0.06 1.6 14 1.29
006* A2-3.17982* B2+0.18695* C 2-0.010943* D2
2 500 0.06 1.2 10 1.46
3 600 0.12 1.4 12 1.28 Where,
4 500 0.1 1.2 10 1.78 A- Spindle Speed
5 700 0.06 1.2 14 1.35 B- Feed Rate
6 600 0.08 1.4 12 1.45 C- Depth of cut
7 700 0.1 1.2 14 1.56 D- End Relief angle
8 600 0.08 1.4 16 1.28
9 600 0.08 1.4 12 1.33
10 500 0.06 1.6 10 1.26 8. Results and Discussion
11 700 0.1 1.2 10 1.78 In this work, the effects of end relief angle, spindle speed,
12 600 0.08 1.4 8 1.48 feed rate and depth of cut were experimentally investi-
13 700 0.06 1.6 14 1.52 gated. The effect of process parameters on the machining
14 700 0.1 1.6 14 1.25 time is discussed below. Figure 3 shows the normal plot of
15 500 0.1 1.6 10 1.52 residuals and Figure 4 shows the Predicted value vs actual
16 600 0.08 1.4 12 1.61 value of machining time.
17 500 0.1 1.2 14 1.6 Figure 5 shows the interaction effect of feed rate and
18 400 0.08 1.4 12 1.68 spindle speed on Machining time. From the figure it is
19 700 0.06 1.6 10 1.32 clearly noticed that the machining time is high between
the range from 400 rpm to 500 rpm and 600 rpm to 800
20 500 0.1 1.6 14 1.64
rpm, whereas the machining time also increase when the
21 700 0.1 1.6 10 1.54
feed rate is increased. Therefore lower feed rate and the
22 600 0.08 1.4 12 1.52
spindle speed between 500 rpm to 600 rpm to be chosen
23 600 0.08 1.8 12 1.29
for the best output result.
24 600 0.06 1.4 12 1.34 From the Figure 6 it influence that the depth of cut
25 700 0.06 1.2 10 1.80 plays a major role for increase in the machining time,
26 800 0.06 1.4 12 1.52 when the depth of cut increases the machining time also
27 600 0.08 1.4 12 1.33 increased. For spindle speed the same result is attained as
28 600 0.04 1.4 12 1.52 shown in Figure 5.
29 500 0.06 1.2 14 1.42 From the Figure 7 it is noticed that at lower end relief
30 600 0.08 1 12 1.39 angle between 8° to 10° the machining time is low whereas
the spindle speed the same result is obtained as shown in
Figure 5.
significant. The Model F-value of 4.72 implies that the From the Figure 8 of interaction diagram the feed
model is significant. There is only a 0.25% chance that an rate and depth of cut increases the machining time also
F-value this large could occur due to noise. The “Lack of increases.
Fit F-value” of 2.97 implies the Lack of Fit is not signifi- From the Figure 9 of interaction diagram the feed rate
cant relative to the pure error. There is a 12.05% chance and end relief angle increases the machining time also
that a “Lack of Fit F-value” this large could occur due to increases.
noise. Non-significant lack of fit is good. The Fig.5 shows From the Figure 10 shows the interaction between
the Predicted Vs Actual model. The regression equation the depth of cut and end relief angle. The depth of cut

Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology 5
 Effect of Machining Parameters and Optimization of Machining Time in Facing Operation using Response Surface Methodology
and Genetic Algorithm

Table 3. ANOVA table for the prediction of Machining Time

ANOVA for Response Surface Quadratic model
Analysis of variance table [Partial sum of squares - Type III]
Sum of Mean F p-value
Source Squares df Square Value Prob > F
Model 0.55 14 0.039 2.67 0.0348 significant
A-Spindle Speed 0.12 1 0.12 8.29 0.0115
B-Feed Rate 6.050E-003 1 6.050E-003 0.41 0.5306
C-Depth of cut 0.011 1 0.011 0.73 0.4055
D-End Relief angle 0.087 1 0.087 5.91 0.0280
AB 0.093 1 0.093 6.34 0.0237
AC 0.042 1 0.042 2.86 0.1113
AD 0.017 1 0.017 1.15 0.3003
BC 0.012 1 0.012 0.82 0.3783
BD 0.024 1 0.024 1.64 0.2203
CD 2.025E-003 1 2.025E-003 0.14 0.7156
A^2 0.084 1 0.084 5.71 0.0304
B^2 6.707E-005 1 6.707E-005 4.568E-003 0.9470
C^2 2.318E-003 1 2.318E-003 0.16 0.6967
D^2 0.079 1 0.079 5.41 0.0345
Residual 0.22 15 0.015
Lack of Fit 0.13 10 0.013 0.66 0.7312 not significant
Pure Error 0.095 5 0.019
Cor Total 0.77 29

Figure 5. Interaction effect feed rate vs spindle speed on machining time.

Figure 5. Interaction effect feed rate vs spindle speed on
Figure 3. NormalFigure
plots
Figure 3. Normal plots of residuals.
of residuals.
3. Normal plots of residuals.
machining
Figuretime.
5. Interaction effect feed rate vs spindle speed on machining time.

Figure 6. Interaction effect depth of cut vs spindle speed on machining time.

Figure 6. Interaction effect depth of cut vs spindle speed on machining time.

Figure 4. Predicted vs actual.

Figure 6. Interaction effect depth of cut vs spindle speed
Figure 4. Predicted vs actual.
Figure 4. Predicted vs actual. on machining time.

6 Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology
 R. Babu, D. S. Robinson Smart, G. Mahesh and M. Shanmugam

Figure 7. Interaction effect end relief angle vs spindle speed on machining time.
Figure 7. Interaction effect end relief angle vs spindle Figure 10. Interaction effect end relief angle vs depth of cut on machining time.
Figure 10. Interaction effect end relief angle vs depth of
speed on machining time. cut on machining time.

9. Evaluation of GA Results
In this present study, the optimization of machining
parameters such as spindle speed, feed rate, depth of cut,
end relief angle was carried out to minimize the machin-
ing time. Genetic Algorithm is used to optimize the
machining time. MAT LAB 7.0 software is used for opti-
mization purpose. The minimization of machining time
by using GA can be expressed by the equation
Figure 11. The performance of fitness value with generation and the best individual
performances of variables in coded form.
Minimize: T (A, B, C, D)
Figure 8. Interaction effect depth of cut vs feed rate on machining time. Within ranges of cutting parameters,
Figure 8. Interaction effect depth of cut vs feed rate on
400 rpm < A < 800 rpm
machining time.
0.04 mm/rev < B < 0.12 mm/rev
Figure 8. Interaction effect depth of cut vs feed rate on machining time. 1 mm < C < 1.8 mm
8° < D < 16° (degree)

To obtain the best optimal results, the number of the

­initial population size, the type of selection function, the
Scaling function, the crossover rate, the mutation rate and
the generations as follows.

Population type: double vector

Figure 9. Interaction effect end relief angle vs feed rate on machining time. Population size: 100
Selection function: Rank
Scaling function: Rank
Function Stochastic: uniform
FigureFigure
9. 9.Interaction
Interaction effect end relief angle vs feed rate on machining time.
effect end relief angle vs feed rate on
Mutation function: Gaussian
machining time.
Mutation rate: 0.1
Crossover function: Scattered
and end relief angle increases the machining time also Crossover rate: 1.0
increases. Generations: 1000
The above interaction diagram it is interesting to
observe that for low machining time, the end relief angle For the machining of the Bimetal bearing, GA predicted
should be in between 8º to 10º. the optimum machining time as 1.169 seconds.

Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology 7
 Effect of Machining Parameters and Optimization of Machining Time in Facing Operation using Response Surface Methodology
and Genetic Algorithm

Table 4. Optimized process parameter predicted by GA

Spindle Speed Feed rate Depth of cut End relief Confirmatory test for Machining Time
Trial (A) (B) (C) angle (D) Predicted GA model Experimental Value % error
rpm mm/rev mm degree µm µm
1 610 0.06 1.4 12 1.61 1.58 1.89
2 700 0.12 1.6 14 1.52 1.49 2.00
3 500 0.08 1.8 10 1.41 1.39 1.43

Figure 10. Interaction effect end relief angle vs depth of cut on machining time.
Machining time compared to the other factors such as
depth of cut and end relief angle.
During machining operation the end relief angle
should be in between the 8°-10°, spindle speed should be
in between the 600 rpm to 800 rpm, feed rate should be
in between 0.04mm - 0.08 mm, depth of cut should be
in 1mm - 1.4 mm. Further the optimization was carried
out on machining time using GA. The predicted result
of machining time was 1.169 seconds of Bimetal bear-
ing. The confirmatory result shows good arguments for
Figure 11. The performance of fitness value with generation and the best individual
experimental vs predicted value.
Figure 11. The performance of fitness
performances of variables in coded value
form. with generation

and the best individual performances of variables in coded

form.
12. References
1. Mahesh G, Muthu S, Devadasan SR. A Review of
Optimization Techniques, Effect of Process Parameter with
10. Validation of the Model Reference to Vibration in End Milling Processes. European
The predicted result of Design of experiments using Journal of Scientific Research. 2012; 76(2):226–39.
2. Qehaja N, Jakupi K, Bunjaku A, Bruci M, Osmani H. Effect
Central Composite Design of RSM and GA is further
of machining parameter and machining time on surface
validated using physical measurements and verified using
roughness in dry turning process. 25 th DAAAM inter-
confirmatory test. The percentage of error is found to be national symposium on Intelligent Manufacturing and
within ±2% which shows the validity of the model. The Automation; 2014. p. 135–40.
experimental results predicted by GA of machining time 3. Abuelnaga AM, El-Dardiry MA. Optimization methods
show good agreement. Table 4 shows the comparison for metal cutting. International Journal of Machine Tool
of predicted vs experimental value of Machining time. Design Research. 1984; 24(1):11–8.
Figure 11 shows the performance of fitness value with 4. Gopalakrishnan B, Fraiz AK. Machining parameter selec-
generation and the best individual performances of vari- tion for turning with constraints: an analytical approach
ables in coded form. based on geometric programming. International Journal of
Production Research. 1991; 29(9):1897–908.
5. Okushima K, Hitomi K. A study of economical machin-
11. Conclusion ing: an analysis of the Maximum profit cutting speed.
International Journal of Production Research. 1964;
The following conclusions has been drawn on facing 3(1):73–8.
operation of a bimetal bearing using Response Surface 6. Raja SB, Baskar N. Application of Particle Swarm
Methodology (RSM) and Genetic Algorithm (GA), con- Optimization technique for achieving desired milled sur-
sidering the machining parameters such as Spindle Speed, face roughness in minimum machining time. Expert
Feed Rate, Depth of cut, and End relief angle using design Systems with Applications. 2012; 39(5):5982–9.
of experiments. The Spindle Speed and Feed rate are the 7. Davim JP. Design of optimization of cutting parameters for
most important parameters to be considered for better turning metal matrix composites based on the orthogonal

8 Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology
 R. Babu, D. S. Robinson Smart, G. Mahesh and M. Shanmugam

arrays. Journal of Materials Processing Technology. 2003; 14. Baskar N, Asokan P, Saravanan R, Prabhaharan G.
132(1-3):340–4. Optimization of Machining Parameters for Milling
8. Rodriguez CA, Harnaut T, Wang Y, Akgerman N, Altan Operations Using Non-conventional Methods. International
T. Estimation of machining time in high speed milling Journal of Advanced Manufacturing Technology. 2005;
of prismatic parts. Proceedings of the North American 25(11):1078–88.
Manufacturing Conference (NAMRC XXVII); Berkeley, 15. Box GEP, Draper NR. Empirical model-building and
California. 1999. p. 142–7. response surfaces. New York: John Wiley; 1987.
9. So BS, Jung YH, Park JW, Lee DW. Five-axis machining 16. Goldberg DE, Holland JH. Genetic algorithms in search,
time estimation algorithm based on machine character- optimization, and machine learning. Netherland: Kluwer
istics. Journal of Materials Processing Technology. 2007; Academic Publishers; 1988; 3(2):95–9.
187–188:37–40. 17. Rouhi F, Effatnejad R. Unit commitment in power system
10. Maropoulos PG, Baker RP, Paramor KYG. Integration of by combination of Dynamic Programming (DP), Genetic
tool selection with design. Part 2. Aggregate machining Algorithm (GA) and Particle Swarm Optimization (PSO),
time estimation, Journal of Material Processing Technology. Indian Journal of Science and Technology. 2015; 8(2).
2000; 107(1-3):135–42. 18. Karthikeyan T, Thangaraju P. Genetic algorithm based
11. Monreal M, Rodrigues AC. Influence of tool path strategy CFS and naive Bayes algorithm to enhance the predictive
on the cycle tine of high speed milling. Computer Aided accuracy. Indian Journal of Science and Technology. 2015;
Design. 2003; 35(4):395–401. 8(26).
12. Coelho RT, de Souza AF, Roger AR, Yoshida RAM, de Lima 19. Hindustan Machine Tools (HMT), Production Technology.
Ribeiro AA. Mechanistic approach to predict real machin- Bangalore: Tata McGraw-Hill Education; 2001.
ing time for milling free-form geometries applying high 20. Palla N. An empirical study of dimensional errors and sur-
feed rate. International Journal of Advanced Manufacturing face roughness in turning. Manhattan, Kansas: Kansas State
Technology. 2010; 46(9):1103–11. University; 2002.
13. Shin YC, Joo YS. Optimization of machining condi- 21. Hoffman EG, McCauley CJ, Hussain MI. Shop reference
tions with practical constraints, International Journal of for students and apprentices. New York: Industrial Press;
Production Research. 1992; 30(12):2907–19. 2000.

Vol 8 (36) | December 2015 | www.indjst.org Indian Journal of Science and Technology 9