See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/225177455

Precision Machining of an Aluminum Alloy

Piston Reinforced with a Cast Iron Insert

Article in International Journal of Precision Engineering and Manufacturing · July 2009

DOI: 10.1007/s12541-009-0002-7

CITATIONS READS

11 75

4 authors, including:

M. Uthayakumar S. Aravindan
Kalasalingam University Indian Institute of Technology Delhi
90 PUBLICATIONS 329 CITATIONS 104 PUBLICATIONS 1,432 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Call for Book Chapters: Green Engineering Techniques for Modern Manufacturing View project

Machining and Erosion studies on Polymer Matrix composites View project

All content following this page was uploaded by M. Uthayakumar on 01 January 2017.

The user has requested enhancement of the downloaded file.

INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1, pp. 7-13 JANUARY 2009 / 7
10.1007/s12541-009-0002-7

Precision Machining of an Aluminum Alloy

Piston Reinforced with a Cast Iron Insert
Marimuthu Uthayakumar1,#, Gopalakrishnan Prabhakaran2, Sivanandham Aravindan3
and Jonna Venkata Sivaprasad4
1 Department of Mechanical Engineering, Kalasalingam University, Krishnan Koil, Tamil Nadu, India, 626-190
2 Department of Mechanical and Industrial Engineering, Caledonian College of Engineering,C.P.O Seeb – 111, Sultanate of Oman
3 Department of Mechanical Engineering, Indian Institute of Technology, New Delhi, 110-016
4 Engineering Department, India Pistons Ltd, Maraimalai Nagar, Chennai, India, 603-209
# Corresponding Author / E-mail: [email protected], TEL: +91-9443918525., FAX: +91-4563289322

KEYWORDS: 3D shape morphing, Template mesh, Implicit function, Mesh smoothing, Modified Laplacian coordinate

Bimetallic pistons consisting of aluminum alloy reinforced with a cast iron (CI) insert are used to reduce the
weight and improve the wear resistance of pistons. A major problem with machining such bimetallic pistons is
producing the desired shape with minimal cutting forces and without damaging the bonding registry. The objective
of this paper is to determine the optimal cutting parameters (cutting speed, feed, and depth of cut) for turning
bimetallic pistons. When machining, we wish to obtain optimal values of the cutting forces and a better surface
integrity while maintaining the required surface finish. Experiments were conducted following Taguchi’s
parameter design approach using a cubic boron nitride tool for the machining. The results indicate that the
process parameters affected the mean and variance of the cutting force at the Al-CI interface of the piston. The Al–
CI interface was examined using an ultrasonic piston bond tester after machining to assure the bond quality. The
surface roughness of the components was measured with a surface roughness tester. A mathematical model was
developed using the Systat 12.0 software package to establish the relationship between the input quantities (speed,
feed, and depth of cut) and the output data (cutting force). The output data of the mathematical model were
compared with the experimental results. The results from the Taguchi robust design concept were compared with
the results obtained from a nonconventional Genetic Algorithm optimization technique.

Manuscript received: January 11, 2008 / Accepted: July 24, 2008

techniques can be used to produce bimetallic components from

NOMENCLATURE
aluminum alloys with CI inserts. In conventional die casting, a liquid
D = depth of cut E = error aluminum alloy is poured into a mold over the insert; the later is
F = feed Fm = mean force simply embedded in the light alloy after solidification. However, an
Fx = feed force Fy = thrust force iron and aluminum oxide layer prevents direct contact from being
Fz = main cutting force n = number of trials established between two metals so that no real interfacial bonding
Ra = average surface roughness S = speed exists. Forming a metallurgical bond at the insert/alloy interface using
S/N or η = signal to noise ratio T = total
a special casting technique derived from the Al-Fin process is
y = response of quality characteristics
possible,1 and the strength of the metallurgical bond can be enhanced
by a suitable heat treatment process.2 This results in a bimetallic
component with improved functionality. Al–CI bimetallic pistons are
1. Introduction generally employed only in diesel engines of heavy-duty vehicles. An
aluminum alloy piston with a CI insert is shown in Figure 1.
To meet weight savings and production cost requirements, many
components for transport vehicles that were previously made of cast
iron (CI) are now being fabricated from light aluminum alloys. These 2. Literature Survey
alloys are often reinforced with a strong metal-based insert, especially
at high load bearing regions, to improve their performance. Different Considerable interest exists in replacing the CI and steel in

© KSPE and Springer 2009

 8 / JANUARY 2009 INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1

automotive components such as pistons with lightweight aluminum optimized are identified for any engineering or nonengineering
alloy castings to improve the performance and efficiency of the optimization problem. Quality characteristics are classified as
vehicle.3 The low expansion group of aluminum-silicon alloy, referred maximation, minimization, or normal criterion. Then the factors
to as piston alloy, provides the best overall balance of properties.4 A (parameters) likely to affect the quality characteristics are identified
metal-based insert reinforces the base alloy to improve the wear and segregated under control (signal) and noise factors. An
performance. Engineering components, such as pistons and the orthogonal array is selected based on the number of parameters and
cylinder head, are first processed by die casting. They are than levels of variations, and a loss function is then defined to calculate the
machined to achieve the desired dimensions and surface finish.5,6 deviation between the experimental and desired values. The selection
of the loss function is based on the quality characteristics. The value
of the overall loss function is further transferred into a signal to noise
ratio (S/N) that is identified by η with units of decibels (dB). The best
value of the identified parameter is predicted from the S/N curves. A
statistical analysis of variance (ANOVA) is performed to find the
percentage effect on the output response. The ANOVA separates the
variation in an experiment into categories related to the cause of the
variation. Finally, a confirmation test is conducted using the best set
of selected values of the identified parameters; this last step is of the
utmost important in the DOE approach.17 The best set of values will
yield a product of superior quality compared to other possible
combinations of values.

4. Experimental Plan
Fig. 1 Typical piston with insert
The identified parameters that affect the characteristics of turned
Metal cutting industries continue to suffer from not being able to
parts are the cutting tool parameters (tool geometry and tool material),
run machine tools at their optimum operating conditions.7 The
workpiece-related parameters (microstructure and hardness), cutting
functional behavior of machined parts is determined by their fine
parameters (speed, feed, and depth of cut), and environmental
finish, which represents the last step in the process chain and can be
parameters (dry cutting and wet cutting). The ranges of the selected
performed by cutting.8 Cubic Boron Nitride (CBN) is a sintered
turning process parameters (cutting speed, feed, and depth of cut)
product that has excellent wear durability, high hardness, and good
were ascertained by conducting preliminary experiments. The details
thermal resistance. It is unlikely to generate a chemical reaction with
of work material, machine tool, cutting tool, and cutting conditions
the workpiece under high thermal conditions, but the cost of CBN
are as follows.
tools is extremely high, between 10 and 20 times greater than the cost
Work material: Aluminum alloy piston reinforced with a CI
of carbide tools.9 The improved qualities of the machined surface and
insert
associated productivity improvement help alleviate their initial high
Machine tool: Leadwell (T5) turning center
cost.
Cutting tool: CBN with a top rake angle of 0°, aclearance
CBN cutting tools are commonly used for single-point turning of
angle of 15°, and a nose of 0.6 mm
hardened materials.10 Due to recent cutting tool developments, as well
Cutting condition: Dry
as the improved machine tool structural rigidity and better CNC
Dynamometer: Kistler Quartz 3 component
controllers, hard turning is emerging as an alternative process to
An orthogonal array gives a more reliable estimate of the factor
grinding due to its reduced cost and cycle time.11 The high CBN
effects with fewer tests compared to traditional methods. The process
content of CBN tools (as much as 90%) gives a better surface finish,
12 parameters along with their values at three levels are given in Table 1.
but the geometry of the cut evolves considerably over the lifetime
of a CBN tool.13 In the past, two different tools and two different
Table 1 Factors and their levels
machining parameters were used to machine bimetallic components.
The cycle time of these processes was unnecessarily long due to the Factors Units Level 1 Level 2 Level 3
frequent tool changes required. Hence, a single tool that can machine
Cutting speed m/min 452 482 512
both metals is desirable. Coated cemented carbide cutting tool inserts
have been used to machine bimetallic engine blocks such as Feed mm/rev 0.15 0.20 0.25
aluminum alloys and CI.14 However, literature on the machining and Depth of cut mm 0.15 0.20 0.25
surface integrity of such bimetallic components is scarce. The present
study addresses current problems encountered when machining
bimetallic components. Three major control factors (cutting speed, feed, and depth of cut)
were selected to conduct the tests. All three factors are multilevel
variables and their outcome effects have nonlinear relationships;
3. Taguchi Method hence, we used three-level tests for each factor. Figure 2 shows the
piston machining arrangement with a dynamometer.
The Taguchi method has produced a unique and powerful quality The number of degrees of freedom was calculated from the
improvement discipline that differs from the traditional process. Since number of parameters identified and their number of levels of
the product and process design has a great impact on the life cycle, variation. Using the full factorial design (3×3×3×3) reduced a total of
cost, and quality of a component, the Taguchi design of experiment 81 sets of experiments down to 9, thereby decreasing the cost, time,
(DOE) approach provides a design engineer with a systematic method and effort.18 The array along with the factor assigned to the columns
for determining the optimum design parameters to obtain the best is presented in Table 2, where L9 indicates the 9 trials considered for
performance at the lowest cost.15,16 The quality characteristics to be the test.
INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1 JANUARY 2009 / 9

developed to estimate the mean cutting force. The results obtained

Piston through the Taguchi method were compared with those obtained
using the nontraditional Genetic Algorithm (GA) method.
CBN Tool

5. Results and Discussion

Spigot 5.1 Cutting force

The machining tests were conducted as per the orthogonal array.
Dynamometer The forces were observed and the mean forces were calculated. The
Attachment experimental results are listed in Table 3.

Table 3 Experimental results

Speed Feed Depth of cut Mean cutting force
Fig. 2 Typical arrangement for machining (m/min) (mm/rev) (mm) at bonding region (N)
452 0.15 0.15 42
Table 2 L9 orthogonal array 452 0.20 0.20 55
Test Column 452 0.25 0.25 82
number Speed (S) Feed (F) Depth of cut (D) 482 0.15 0.20 60
1 1 1 1 482 0.20 0.25 85
2 1 2 2 482 0.25 0.15 58
3 1 3 3 512 0.15 0.25 53
4 2 1 2 512 0.20 0.15 39
5 2 2 3 512 0.25 0.20 76
6 2 3 1
7 3 1 3 The cutting forces present during machining were measured using
a dynamometer. A typical graphical representation of the cutting force
8 3 2 1
(Fz) while machining the bimetallic piston is shown in Figure 3. The
9 3 3 2
observed cutting force while machining the aluminum alloy portion
of the bimetallic piston was low, around 30 N. The cutting force
Pistons that were 96 mm in diameter and 30 mm in length (with
increased to around 55 N when the tool started to machine the harder
bonding zones from one end) were machined on a turning center with
CI material. The subsequent reduction in the cutting force profile was
a CBN tool. The cutting forces were measured using a dynamometer
due to a groove in the CI.
while each specimen was turned. Dynamometers have been
To find the optimal cutting conditions, the selected quality
developed that are capable of measuring tool forces with an
characteristic (cutting force at bonding zone) should be of lower
increasing degree of accuracy.19 The component forces were also
order; hence, the S/N for the “lower the better” type of response was
taken into account and the mean force was calculated from the
used:
formula

Fm = Fx 2 + Fy 2 + Fz 2 (1)
1
(
S/N ratio (in dB) = − 10log  ∑ y12 + y22 + ......... yn 2
n
) , (2)

The optimal cutting conditions were identified from the

experimental results by applying the ANOVA. The surface integrity where y is the response of the quality characteristic for the trial
of the machined piston was evaluated through bond checking and condition and n is the number of trials. The S/N was computed for all
surface roughness measurements. A mathematical model was then trials; the results are listed in Table 4.

80 Fz [N]
Cutting Force (N)

60

40

20

0
5 10 15 20 25 30
-20
Time (s)
-40
-60

Fig. 3 Typical cutting force variation

 10 / JANUARY 2009 INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1

Table 4 Computing the cutting force variation

Speed Feed Depth of cut
Factors Error Total
(m/min) (mm/rev) (mm)
1 14.82 15.73 16.82 13.78
Sum at factors levels 2 13.43 14.93 13.91 15.14 43.82
3 15.57 12.94 12.88 14.68
Sum of the square of the difference(s) 7.07 12.38 25.05 2.87 47.37
Contribution ratio (%) 14.93 26.13 52.88 6.06 100

Table 5 Variation of cutting force

Source of variations Sum of the squares Degrees of freedom Mean sum of the squares Variance
Speed 7.07^ 2 3.54 2.46
Feed 12.38** 2 6.19 4.30
Depth of cut 25.05* 2 12.53 8.70
Error 2.87 2 1.44
Total 8
If the variance value is greater than 2, it is considered a significant term.
* indicates very high significance.
** indicates significance.
^ indicates the least significance.

Table 6 Cutting force contribution

Factor Depth of cut Feed Speed Error
Contribution rate (%) 52.88 26.13 14.93 6.06
Cumulative contribution rate (%) 52.88 79.01 93.94 100

The data analysis procedure using the Taguchi experimental values of the cutting force for each parameter at levels 1, 2 and 3 were
framework involves an analysis of means and variances. The analysis calculated and plotted in Figure 5. The effect of the factor level was
of means helps to identify the factor contributions whereas the defined as the deviation it caused from the overall mean expressed in
ANOVAs establish the relative significance of the factors in terms of decibels. This process of estimating the main effects of each factor is
their percentage contribution to the response. known as an analysis of the mean. The minimum cutting force
occurred at the 3rd speed level, 1st feed level, and 1st depth of cut level,
5.2 ANOVA scheme for an orthogonal array experiment giving [3 1 1] as the best level to obtain the minimum cutting force
The ANOVA scheme was used to study the significance of the for the piston.
parameters affecting the quality characteristics of interest. The
scheme subdivides the total variation in the data into useful and
meaningful components of variation. The total variation contribution S/N RATIO CHART
is shown in Table 5.
O
IT
20

The ANOVA results in Table 6 clearly identify that the depth of

AR S3 F1
D1

cut and feed severely affected the cutting force by 52.88% and N/
15 S1
S2
F2
F3
D2
D3

26.13%, respectively, while the speed only affected the cutting force S
E 10

by 14.93%. A Pareto chart generated based on the contribution ratio is G

AR
EV
5
presented in Figure 4. This chart shows the importance of the
significant parameters. A 0

S1 S2 S3 F1 F2 F3 D1 D2 D3

PARAMETER
PARETO CHART

O
IT 60 Fig. 5 Effects of the factor level on the cutting force
AR 50
N 40
O Conducting a verification experiment is a crucial and the last step
TIU 30
BI 20 of the robust design procedure. Its aim is to verify the optimum
RT conditions identified by the matrix tests and it estimates how close the
N 10
O
C 0 predictions are to actual conditions. Hence, a confirmation test was
Depth of1cut Feed 2 3
Speed
conducted with the optimum parameters, with good results.
PARAMETERS
5.3 Mathematical model
Fig. 4 Pareto chart The design optimization method makes extensive use of
functional or empirically developed mathematical models that
The mean responses refer to the average value of the performance explicitly link a quantitative dependent variable to certain
characteristics for each parameter at different levels. The average independent variables.20 The time-tested approach to building such
INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1 JANUARY 2009 / 11

models from observed quantitative data is a regression analysis. In the

Initialize Parameters
present study, empirical models were developed to approximate the
cutting force using the Systat 12.0 software package based on a
regression analysis of the experimental results. The cutting forces Generate Initial Population
predicted using the developed model compared are compared to
experimental values in Figure 6. Little variation was observed
between the modeled and experimental values, confirming the Evaluate Fitness Function for Entire Population
applicability of the mathematical model.
Evaluate Population Statistics
Comparison chart
90 Generate Parents Update
80 Population
)
N
(
70
e 60
Experimental values
c
r
o
f
50 Perform Crossover
g 40 Model values
30
n
i
t
t
u
C 20
10 Generate Offspring with Mutations
0
1 2 3 4 5 6 7 8 9
Experimental Run

Fig. 6 Comparison between model and experimental results Is the

Optimal
5.4 Genetic algorithm Solution
The GA is a search strategy that is able to search a large solution
space efficiently by providing a concise computational cost because it
uses probabilistic transaction rules instead of deterministic ones.21 It
is easy to implement and is increasingly being used to solve
inherently intractable problems quickly. A GA is a computerized Terminate
search and optimization algorithm based on the mechanics of neural
genetics and natural selection.22 The quality is the fitness function, Fig. 7 GA flowchart
which evaluates a chromosome23 with respect to the objective
function of the optimization problem. Using genetic operators Table 7 Comparison between Taguchi results and GA output
(crossover and mutation), the algorithm creates a subsequent Optimum parameters Taguchi method GA output
generation from a string of the current population. The generation
Speed (m/min) 512 506
cycle is repeated until a desired termination criterion is reached. A
simple genetic algorithm that yields good results for many practical Feed (mm/rev) 0.15 0.1517
problems has three main operations: Depth of cut (mm) 0.15 0.15329
• reproduction,
• crossover, and
bond zone between the aluminum and CI was checked using an
• mutation.
ultrasonic bond-checking machine (EPST II). The pistons were
Reproduction is a process in which individual strings are copied
placed in such a way that the two probes were touching the top and
according to their objective function values. Crossover is the process
bottom portions of the bonding region. The piston was rotated and the
of exchanging of information between or among the mating string. In
probe sensed if a debonding zone was present on the circumference of
most crossover operators, two strings are picked from the mating pool
the Al–CI interface. Observations were made from the display graph
at random, and some portions of the strings are exchanged between
and from the instantaneous signal from the controller. The bond-
the strings. Mutation is the occasional random alteration of the values
checking facility is shown in Figure 8.
in a string, and serves to create a point in the neighborhood of the
current point, thereby achieving a local search and the current
solution. The GA can be used for general optimization problems. A 21.12.2006

typical flowchart for a GA is presented in Figure 7.

The mathematical model created using the Systat 12.0 software
package to relate the cutting force and process parameters was
encoded for the Matlab software package in M-file format. Then the
GA Toolbox was used to input the fitness function, given as
@filename. The number of variables was 3, and other GA
functionalities were chosen accordingly to obtain the best fitness
value in a random process. The best fitness value obtained from the
GA is compared with Taguchi’s optimum parameters in Table 7.

6. Surface Integrity EPST - 11 ULTRASONIC PISTON BOND TESTER

ELEVATION -1
6.1 Bond checking
The integrity of all the machined surfaces of the pistons at the Fig. 8 Ultrasonic bond-checking facility
 12 / JANUARY 2009 INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1

6.2 Surface roughness 7. Summary and Conclusions

The surface quality of a machined component is one of the most
specified customer requirements. A major indication of the surface The following summary and conclusions can be drawn from our
quality on a machined part is the surface roughness, which is mainly a experiment.
result of the process parameters such as the tool geometry and cutting 1. A CBN tool was used to machine bimetallic pistons.
conditions. A typical surface roughness measuring facility, the 2. The optimal cutting conditions for bimetallic pistons were
Mitutoyo CV2000, is shown in Figure 9(a), and typical results given obtained using the Taguchi method.
by the display screen are presented in Figure 9(b). The piston was 3. An ANOVA analysis indicated that the depth of cut and feed rate
placed over the surface plate and the probe was moved on the were the most important parameters that affected the cutting force.
machined surface at predetermined intervals (30 mm) to obtain the 4. The developed mathematical model accurately predicted the
average surface roughness. These measurements were performed on cutting force at the bonding zone.
all the machined surfaces. 5. The optimal results obtained using the Taguchi method were
The average surface roughness (Ra) value of all the machined comparable with the results obtained using a GA.
pistons is shown in Figure 10. Normally, the maximum allowable 6. The surface integrity of the machined pistons was evaluated using
surface roughness for a piston is 4 µm. All the tested pistons had surface roughness and bond-checking tests. The machined pistons
satisfactory roughness values, and the empirical trials indicated that a exhibited good surface integrity for the chosen level of parameters.
lower surface roughness resulted when the feed rate was low.

ACKNOWLEDGEMENTS

The authors submit their humble thanks to the management of

India Pistons Ltd, Chennai for their technical support and the
rendered permission to publish in this journal.

REFERENCES

1. Viala, J. C., Peronnet, M., Barbeau, F., Bosselet, F. and Bouix, J.,
“Interface Chemistry In Aluminum Alloy With Iron Base Inserts,”
Composites, Part A, Vol. 33, No. 10, pp. 1417- 1420, 2002.

2. Uthayakumar, M., Prabhaharan, G., Aravindan, S. and Sivaprasad,

(a)
J. V., “Study on aluminum alloy piston reinforced with cast iron
insert,” International Journal of Materials Science, Vol. 3, No. 1,
pp. 1-10, 2008.

3. Cole, G. S. and Sherman, A. M., “Light weight aluminum alloy

for automotive application,” Material Characterization, Vol. 35,
pp. 3-9, 1995.

4. Haque, M. M. and Sharif, A., “Study on wear Properties of

aluminum-silicon piston Alloy,” Journal of Material Processing
Technology, Vol. 118, No. 1-3, pp. 69-73, 2001.

5. Durrant, G., Gallerneault, M. and Cantor, B., “Squeeze cast

aluminum reinforced with Mild steel inserts,” Journal of Material
Science, Vol. 31, No. 3, pp. 589-602, 1996.
(b) 6. Dwivedi, D. K., Sharma, A. and Rajan, T. V., “Machining of LM
Fig. 9 (a) Surface roughness measurement facility and (b) typical 13 and LM 28 Cast aluminum alloys : Part I,” Journal of
displayed results Materials Processing Technology, Vol. 196, No. 1-3, pp. 197-204,
2008.
Surface roughness chart 7. Singh, H. and kumar, P., “Optimizing cutting force for turned
parts by Taguchi’s parameter design approach,” Indian Journal of
4
Engineering Materials, Vol. 12, No. 2, pp. 97-103, 2005.
S urface roughness value

3
8. Tonshoff, H. K., Arendt, C. and Ben Amor, R., “Cutting of
hardened steel,” CIRP Annals Manufacturing Technology, Vol. 49,
(Ra)

2
Issue 2, pp. 547-566, 2000.
1
9. Lin, Z. C. and Chen, D. Y., “A study of cutting with CBN tool,”
0 Journal of Material Processing Technology, Vol. 49, No. 1-2, pp.
1 2 3 4 5 6 7 8 9
149-164, 1995.
Experimental Run
10. Liang, S. Y. and Huang, Y., “CBN tool wear modeling in finish
Fig. 10 Surface roughness measurements hard turning,” ASME International Mechanical Engineering
INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING Vol. 10, No. 1 JANUARY 2009 / 13

Congress and Exposition, IMECE2003-42165, 2003.

11. Buyukhatipoglu, K., Lazogliu, I., Gratz, H. and Klocke, F.,

“Mechanics and dynamics of hard turning process,” ASME
International Mechanical Engineering Congress and Exposition,
IMECE2004-61067, 2004.

12. Pavel, R., Marinescu, I. D., Sinram, K. and Deis, M., “PCBN
tools wear in continuous and interrupted hard turning,” ASME /
STLE Joint International Tribology Conference, TRIB2002-127,
2002.

13. Mehdi Remadna, M. and Rigal, J. F., “Evaluation during time of

tool wear and cutting forces in the case of hard turning with CBN
inserts,” Journal of Materials Processing Technology, Vol. 178,
No. 1-3, pp. 67-75, 2006.

14. Hessman, I., “Cutting tool for bimetal machining,” U.S. Patent,
No. 20050181211.

15. Roy, R., “A Primer on the taguchi method,” Van Nostrand, pp. 11-
28, 1990.

16. Ross, P. J., “Taguchi technique for quality engineering,”

McGraw-Hill, pp. 92-105, 1998.

17. Bagchi, T. P., “Taguchi methods explained practical steps to

robust design,” Prentice Hall of India (p) Ltd., pp. 42-54, 1993.

18. Gaitonde, V. N, Achyutha, B. T. and Siddeswarappa, B., “Burr

size minimization in drilling using Taguchi technique,” Indian
Journal of Engineering & Material Sciences,” Vol. 12, No. 2, pp.
91-96, 2005.

19. Sharma, V. S., Dhiman, S., Sehgal, R. and Sharma, S. K.,

“Assessment and Optimization of Cutting Parameters while
Turning AISI 52100 Steel,” International Journal of Precision
Engineering and Manufacturing, Vol. 9, No. 2, pp. 54-62, 2008.

20. Palanikumar, K. and Davim, J. P., “Mathematical model to

predict tool wear on the machining of glass fibre reinforced
plastic composites,” Materials & Design, Vol. 28, Issue 7, pp.
2008-2014, 2007.

21. Dereli, T. and Filiz, I. H., “Optimisation of process planning

functions by Genetic algorithm,” Computers and Industrial
Engineering, Vol. 36, Issue 2, pp. 281-308, 1999.

22. Kalyanmoy, D., “Optimisation for engineering design algorithm

and example,” Prentice Hall of India (p) Ltd., pp. 1-51, 2000.

23. Saravanan, R., Asokan, P. and Sachithanandam, M.,

“Comparative analysis of conventional and non conventional
optimization techniques for CNC turning process,” International
Journal of Advanced Manufacturing Technology, Vol. 17, No. 7,
pp. 471-476, 2001.

View publication stats