Journal of Cleaner Production 66 (2014) 309e316

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Optimization of process parameters using a Response Surface Method

for minimizing power consumption in the milling of carbon steel
Gianni Campatelli*, Lorenzo Lorenzini, Antonio Scippa
University of Firenze, Department of Industrial Engineering, Via di S. Marta 3, 50139 Firenze, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Due to the urgent need for global reductions of environmental impacts, many studies have been carried
Received 17 January 2013 out in different fields. One of the most important sectors is manufacturing, particularly due to the high
Received in revised form power consumption of the production machines of manufacturing plants. This paper focuses on the
10 October 2013
efficiency of the machining centres and provides an experimental approach to evaluate and optimize the
Accepted 14 October 2013
process parameters in order to minimize the power consumption in a milling process performed on a
Available online 26 October 2013
modern CNC machine. The parameters evaluated are the cutting speed, the axial and radial depth of cut,
and the feed rate. A lubrication strategy has been chosen based on previous studies: all the tests have
Keywords:
Green manufacturing
been carried out using dry lubrication in order to eliminate the environmental impact due to lubricant
Parameter optimization without substantially affecting the energy consumption. The process has been analyzed using a Response
Milling Surface Method in order to obtain a model fit for the fine tuning of the process parameters.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction efficiency could also be obtained by developing more efficient

One of the most relevant challenges in modern manufacturing is
the need to reduce the environmental impacts of production as far
as possible. Often the production companies seek to change dras-
tically their technologies in order to adopt a greener approach.
Unfortunately this is not a choice for every product: sometimes
alternative and greener technologies for manufacturing a specific
component do not exist and thus it is necessary to implement an
incremental improvement of the current technology. This is often
the case for products manufactured by milling, which usually have
strong constraints due to the required surface finish and geomet-
rical complexity which cannot be obtained using alternative
technologies.
The EU Commission recognized that machine tools play a very
important role in this scenario. Their efficiency is so crucial that this
type of product has been proposed for inclusion in the product
categories regulated by the Ecodesign Directive (EPTA, 2007). Also,
the Kyoto Protocol (United Nations Framework Convention on
Climate Change, 1997) reports on the importance of achieving re-
ductions in energy use through the development of different design
features of machine tools by machine tool designers. Another
crucial issue is related to machine tool users: an increase in
* Corresponding author. Tel.: þ39 (0)55 4796291; fax: þ39 (0)55 4796400.
E-mail address: gianni.campatelli@unifi.it (G. Campatelli).
machine tool motions and tool paths.
Moreover, interest in using green manufacturing strategies is
steadily increasing in manufacturing companies, mainly due to two
reasons: the environmental cost of production is taken into account
by an ever-increasing number of governments and depletion of
world resources is occurring so material and energy costs are
increasing (i.e. the trend in the cost of iron is very well known by
mechanical companies). With these boundary conditions it is
necessary to focus the research efforts in order to create greener
manufacturing strategies.
The objectives of green manufacturing are summarized well by
the 12 principles of Anastas and Zimmermann (Anastas and
Zimmerman, 2004); in particular the science of manufacturing
has the task of reducing the environmental impact of its processes
through parameter optimization and/or changes of technologies
and materials.
In the product design field, many studies have been carried out
in order to evaluate the environmental impact of certain materials
or technical solutions, thanks to the development of structured and
database-based tools such LCA (Life Cycle Assessment, defined by
the ISO 14000 standards) followed by LCM (Life Cycle Manage-
ment). These analyses add new value to the product, helping the
diffusion of ISO 14001 certification among production companies;
the feasibility and advantages are already well documented
(Orecchini, 2000; Caspersen and Sorensen, 1998). However in these
analyses the technology plays a very unimportant role: most of the
databases offer only a crude evaluation of the environmental

0959-6526/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jclepro.2013.10.025
310 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316
impacts of a specific technology (i.e. the Ga.Bi. database), driving
the designer to choose a new technology instead of providing a tool
for the optimization of the current one. In Europe (2009 data from
Europe’s Energy Portal, 2012), about 24% of the primary energy
consumption is used by the industry and a margin for improvement
of about 30% has been estimated. Fig. 1 shows the Italian 2010
power consumption data by sector, while Fig. 2 presents a detailed
analysis of the industrial sector; the data reported in these figures
have been available on the website of the Italian energy distributor,
TERNA (2010 data from TERNA website, 2012). From these data the
significant contribution due to machining is evident.
Many actions could be carried out in order to reach this goal: the
development of zero defect processes, the implementation of net
shape processes, the use of reusable materials, and the reduction of
resource consumption by production machines. While the first
actions require very consistent investment in order to drastically
change production technologies, the last approach could be applied
easily in order to maintain the same production system while
simply optimizing its process parameters. Moreover, optimizing
existing processes instead of making drastic changes has an
intrinsically better economic and social sustainability due to the
lower investment needed and user acceptance (Pusavec et al.,
2010a, 2010b).
2. Analysis of the environmental impact of machining
The demand for environmentally friendly processes imposes a
new perspective on the analyses carried out till now: one example
is the study of the optimal lubrication strategy considering not only
the surface finish or process stability but also new indicators such
as the power consumption and the global environmental impact of
such a manufacturing setup. Many authors have developed models
to assess the environmental impact of machining operations and to
fine tune this value. In order to obtain this result the first step is to
develop a model that allows analytical analysis of the effect of
process parameters on the environmental performance of the
process itself.
In particular many studies have focused on the effect of cutting
fluids on the environment (Lawal et al., 2013; Pusavec et al., 2010a,
2010b; Zackrisson, 2005) and alternative hybrid processes
(Neugebauer et al., 2012). Coolant and lubricants are usually
employed in machining operations, including turning, milling,
drilling, or grinding, and they are able, in general, to improve
machining performances, but there are also some risks associated
with their use (Weinert et al., 2004; Sokovic and Mijanovic, 2001;
Cetin et al., 2011): these fluids are responsible for skin and
breathing problems among machine operators. Furthermore, after
their disposal, if recycling is not possible, they may become
polluting agents in soil and water when inappropriately handled
(Shashidhara and Jayaram, 2010). However a general result could
Fig. 1. Power consumption by sector in Italy (year 2010).
Fig. 2. Power consumption in Italy (year 2010) for the industrial sector in GWh.
be obtained from all these studies: flooded lubrication is an
approach that induces a large environmental impact, and thus the
route taken to reduce the environmental impact must lead to dry or
nearly dry approaches, like MQL (Minimal Quantity Lubrication). In
a previous work by the author (Campatelli, 2009) the effect of
lubrication in a turning process was studied and its environmental
impact was analytically evaluated in terms of both equivalent GWP
(Global Warming Potential) and power consumption due to the
accessory equipment (i.e. cutting fluid pump). The general result
was that the most environmental friendly solution for cutting is dry
lubrication, although MQL is also able to obtain good performance
(þ8% of the environmental burden with respect to dry lubrication).
Moreover the use of dry cutting is encouraged by the reduction of
the processing cost; it has been estimated that the cost of cutting
fluids is approximately 7e17% of the total cost of the machining
process (Klocke and Eisenblaetter, 1997). For newer milling tools for
traditional materials, this saving is able to compensate for the
greater cost of tooling due to the increase in tool wear, which is
reduced by current high performance coatings. Thanks to these
characteristics, dry machining is an emerging trend for traditional
materials such as aluminium (Fratila and Caizar, 2011, Dhar et al.,
2007) or low alloyed steel (Campatelli, 2009) but it can hardly be
used for difficult-to-cut materials, for which MQL is becoming the
best option (Devillez et al., 2007). Another advantage of dry
machining is that the implementation of such an approach does not
require structural changes in machine tools but only some adjust-
ments regarding the tooling and cutting conditions.
Given the previous issues the choice for the experimental tests
carried out was the use of dry machining.
Also, many studies have already been developed regarding the
modelling of the machining processes, sometimes using analytical
approaches and sometimes experimentally based ones. An
important contribution has been the proposal of a standard defi-
nition of the machine tool state in order to have a taxonomy of the
power consumption; in this field the CO2PE! Project (Cooperative
Effort in Process Emission) proposed a unified taxonomy
(Ostaeyen, 2010) and methodology (Kellens et al., 2012) so that
energy data collection in manufacturing can be standardized and
presented in a globally compatible approach. The two states are
‘basic state’ and ‘cutting state’; this taxonomy allows some useful
considerations to be highlighted in order to provide a functional
model for machine tool power consumption. Some of these are
reported in the work of Balogun and Mativenga (2013) and are
briefly summarized here. The basic state power consumption
comprises a higher percentage of total power consumption in
modern machines compared to older machines, as shown by the
work of Kordonowy (2001), Gutowski et al. (2006) and Dahmus
and Gutowski (2004), where the fraction of power used by the
 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316
basic state with respect to the total power shifted from 51.9% to
85.2% when moving from an older to a newer machine. The reason
for this is the ever-increasing presence of support systems, which
are necessary to achieve better performance in terms of cutting
speed and feed, such as more powerful and sophisticated motor
control, lubrication strategies, cooling systems, and so on. Also the
introduction of high accuracy machines with greater complexity
and higher mass to be moved has the result to increase the total
energy consumption in a production shift, as presented by the
works of Diaz et al. (2009) and Vijayaraghavan and Dornfeld
(2010), both of which studied the same Mori Seiki three-axis
NV1500DCG machine. This trend is also confirmed for five-axis
machines, where generally the weight of the structure is heavier
than that of three-axis machines due to the presence of additional
motors and support systems (Devoldere et al., 2007). On the other
hand simpler machines like lathes are more efficient in terms of
the percentage of power used for the cutting with respect to the
power used to support the basic state. The percentage of cutting
power in this case becomes 61e69% for a modern CNC lathe
(Rajemi et al., 2010).
Regarding the modelling of the effect of process parameters,
many authors have developed studies based on different materials
and have used different metrics to evaluate the power consumption
(He et al., 2012; Li and Kara, 2011; Draganescu et al., 2003). To
model the machine consumption, different machine states are also
introduced sometimes; one example is the definition of the ready
state together with the basic and cutting states by Mori et al. (2011),
which is useful to introduce a more detailed study of the power
consumption of the auxiliary systems during the non-cutting
condition. This analysis is not of interest from the machine user’s
point of view but it could be extremely interesting for machine tool
manufacturers in order to plan the optimization of the machine
components and activation strategies. Regarding the metric used,
most authors prefer to analyze the power consumption using the
power needed instead of the energy needed for the removal of a
specific quantity of material; however these two measures are al-
ways comparable. Most of the models proposed are mainly similar
and are composed by the power consumption during the two or
three machine states modelled, like in (1) (Mori et al., 2011), where
P1, P2, and P3 are the power need in basic, idle, and cutting machine
conditions, respectively, while T1, T2, and T3 are the timespans of
machine state activation and Edir is the total direct energy
requirement.
Edir ¼ P1  ðT1 þ T2 Þ þ P2  T2 þ P3  T3
Special attention can be paid to the equation used by Diaz et al.
(2011), which links the specific energy need (ecut) to the Material
Removal Rate (MRR) of the machining process by introducing two
constants (k and b, respectively the constant of instantaneous
specific cutting energy and the specific cutting energy contribution
due to the basic state consumption of the machine) that define the
specific power consumption of the cutting and steady states,
respectively (2).
1
ecut ¼ k  þb
MRR
This model allows explicit reference to MRR to be made to
measure the efficiency of the process. The models developed are
useful for measuring the machining efficiency and rely on experi-
mental data in order to provide a detailed analysis. This means that
the experimental tests connected to a power optimization process
must be extremely accurate and would allow a fine evaluation of
the effect of process parameters on the power consumption in
different machine states.
311
3. Experimental tests
The present study analyzes the effect of simultaneous variations
of four cutting parameters (cutting speed, feed rate, and radial and
axial depth of cut) on energy consumption. For this purpose, the
Response Surface Method (RSM) is utilized. RSM is a group of
mathematical and statistical techniques that are useful for
modeling the relationship between the input parameters (cutting
conditions) and the output variables (energy consumption)
(Montgomery, 2001). RSM saves cost and time in metal-cutting
experiments by reducing the overall number of tests required. In
addition, RSM helps by describing and identifying, with great ac-
curacy, the effect of the interactions of different independent var-
iables on the response when they are varied simultaneously (Mead
and Pike, 1975; Hill and Hunter, 1966; Hicks, 1993). Moreover the
RSM approach enables the factor’s optimum levels to be estimated
much more accurately; it is possible not only to evaluate which of
the levels considered is the best but also to find the exact value that
optimizes the design (Fnides et al., 2011; Horng et al., 2008). The
price of this fine adjustment is a greater number of experiments
than in the other test plans and the requirement that only contin-
uous factors can be used. These features of this approach have
promoted its use in other machine tool energy optimization studies
like the one by Mori et al. (2011). Similar approaches have also been
used for studying the optimization of cutting fluid parameters
(Kuram et al., 2013; Fratila and Caizar, 2011), power consumption,
and tool life (Bhushan, 2013). The formula used for the regression of
the experimental data is quadratic and can be expressed by the
following Equation (3), considering n variables (xi.xn):
X
n X
i X
n
Performance ¼ aij $xi $xj þ bi $xi þ c (3)
i¼1 j¼1 i¼1
Where aij, bi and c are the constants of the equation to be deter-
mined using a regression approach.
The advantage of RSM is given also but the non linear formu-
lation, in fact most of the traditional design of experiments ap-
proaches uses a linear model to approximate the process, that
improve the modelling accuracy. RSM has been extensively used in
the prediction of responses such as tool life, surface roughness, and
cutting forces. Noordin et al. (2004) used the RSM to investigate the
tangential cutting force in the turning of AISI 1045. They found that
the feed rate, as a main factor, and the side cutting edge angle, as a
(1) secondary factor, affected the response variable (tangential force).
The test plan that has been adopted is a central composite one with
an alpha factor equal to two, which provides rotatability, and the
spherical design. The advantage of a 5 level and 4 factors fractional
central composite RSM is the possibility to reduce the number of
tests to be carried out. In particular, using a fractional reduction for
the cube points of the test plan, is possible to reduce the number of
test configuration to 31, starting from the 625 possibilities that
would be computed for a full factional test plan. Three replications
of the tests have been randomly carried out in the same day
(2)
Table 1
RSM factors and levels.
Levels Factors
vc (m/min) ft (mm/tooth) ae (mm) ap (mm)
1 60 0.070 0.8 6
2 70 0.085 0.9 7
3 80 0.100 1.0 8
4 90 0.115 1.1 9
5 100 0.130 1.2 10
312 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316

without relevant thermal transients to be considered. In the Table 1

are shown the factors included in the experimental test plan. The
obtained test plan is reported in Table 2.
During each test the total electrical power consumed by the
machine tool was acquired using three current gauges and voltage
meters whose signal was acquired with an NI-DAQ simultaneous
sampling card, NI-9215, at the frequency of 10 kHz/channel. The
signals were post-processed using a MatlabÒ routine in order to
acquire the instantaneous power during each test. The machine
tool used for the experimental activity was a high performance
five-axis milling machine produced by Mori Seiki, the
NMV1500DCG loaned to the authors by MTTRF for research activ-
ities. This machine is equipped with a high speed spindle that is
able to reach 40,000 rpm and its maximum power is 5.5 kW. The
working space is limited to 420 mm  210 mm  400 mm
respectively in the x, y and z direction. The toolholder is an HSK 32 E
with an elastic collet. The tool used was a carbide one produced by
Osawa (G2CS2 series) with a 10 mm diameter, two cutting edges,
and a monolayer PV200 surface coating. The tool has a helix angle
of 30 and is specially designed for medium hardness steel, it has a Fig. 3. Experimental setup for the tests.

rake angle of 5 , a design tooltip radius of 5 mm and a normal

profile. The total length of the tool was 75 mm while the height of
edges that usually last for the very first seconds of use of the tool
the cutter was 25 mm. The material used for the tests was AISI 1050
and affect the cutting forces. The duration of usage of each tool did
carbon steel with a measured hardness of 247 HB.
not exceed 3 min, which was enough to execute all 31 runs.
For each test a length of material of 100 mm was machined
The acquisition of the power consumption was designed to
(Fig. 3) with the cutting parameters defined in Table 2. The tests
make it possible to obtain as much information as possible
were carried out in a random order to reduce the risk of systematic
regarding the power consumption of the machine components, and
errors in the results due to tool wear. In particular, the tool wear
multiple machine states were defined in order to post-process the
was measured at the beginning and end of the test campaign and,
obtained data accurately. In particular, for each test the data
due to the reduced quantity of total material removed, negligible
acquisition started with the tool standing still and the machine in
signs of wear are reported. Before starting the test campaign, the
idle state (red in Fig. 4). Then the spindle was powered up and the
tool was used for one minute with the process parameters provided
quantity of energy needed for the operation was recorded (orange
by the manufacturer in order to avoid the effects of new sharp
in Fig. 4). By using the G04 ISO command that impose a fixed
waiting time between machine operations is possible to acquire the
Table 2
stand-by power consumption of the machine in a stable state. The
RSM central composite test plan. following operation is the movement of the axis at the cutting feed
(green in Fig. 4). In these experimental tests, rapid movements of
Run Cutting speed Radial engagement Feed per tooth Depth of cut
(vc in m/min) (ae in mm) (ft in mm/tooth) (ap in mm)
the axes were avoided in order to reduce as possible the energy
consumption due to the high acceleration required for such oper-
1 70 0.9 0.085 9
ations. However the effects of the axis acceleration and the path
2 90 0.9 0.085 9
3 70 1.1 0.085 9 definition in order to save energy have been presented by the au-
4 90 1.1 0.085 9 thors in another paper (Campatelli, 2013). Finally the yellow zone
5 70 0.9 0.115 9 represents the energy consumption due to the machining process
6 90 0.9 0.115 9
and the purple one is related to the braking of the spindle until the
7 70 1.1 0.115 9
8 90 1.1 0.115 9
return to idle condition. In Fig. 5 a graph of the idle and cutting
9 70 0.9 0.085 11 power is shown. The power consumption due to the use of com-
10 90 0.9 0.085 11 pressed air and the chip conveyor has been excluded from this
11 70 1.1 0.085 11 analysis; they are, respectively, powered up by a different system or
12 90 1.1 0.085 11
turned off during the tests.
13 70 0.9 0.115 11
14 90 0.9 0.115 11
15 70 1.1 0.115 11
16 90 1.1 0.115 11 4. Results and discussion
17 60 1 0.1 10
18 100 1 0.1 10 The results of the experimental tests have been studied in order
19 80 0.8 0.1 10 to measure the machining power and the total power needed for
20 80 1.2 0.1 10
21 80 1 0.07 10
the material removal process. The results have been normalized
22 80 1 0.13 10 considering the energy needed for the removal of a specified
23 80 1 0.1 8 quantity of chip. The results are expressed in terms of:
24 80 1 0.1 12
25 80 1 0.1 10
 E: energy (J) needed for the operation by the spindle and axis
26 80 1 0.1 10
27 80 1 0.1 10 only;
28 80 1 0.1 10  Es: specific energy (J/mm3), referring to the energy needed by
29 80 1 0.1 10 the axis and spindle to remove 1 mm3 of material;
30 80 1 0.1 10  E tot: total energy (J), considering all the consumption sources of
31 80 1 0.1 10
the machine as a whole;
 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316 313

Fig. 4. Analysis of the power consumption during a machining test.

Fig. 5. Idle and cutting power consumption for a machining test.

 Es tot: total specific energy (J/mm3), referring to the total energy Table 3
consumed to remove 1 mm3 of material. Results of the test campaign.

Run E (J) Es (J/mm3) E tot (J) Es tot (J/mm3)

The results are summarized in Table 3.
1 6663 8.23 39,245 48.5
From this analysis it is possible to evaluate the effect of process
2 4444 4.44 24,462 30.2
parameters on both the efficiency of the cutting process, which is 3 8019 8.1 40,159 40.6
mainly related to the specific energy, and the overall efficiency of 4 5736 5.79 30,675 31.0
the machine, which is related to the total specific energy. In this 5 5990 7.4 29,713 36.7
case the contribution of the machine is significant because the end 6 4299 5.31 22,829 28.2
7 5856 5.92 26,154 26.4
mill used has a small diameter and the related power consumption
8 4948 5 23,411 23.6
is comparable with the power used for the auxiliary system. Using 9 7508 7.58 39,593 40.0
these parameters, the specific energy is about one quarter of the 10 5244 5.3 30,215 30.5
total specific energy, in accordance with the initial considerations 11 9084 7.51 41,240 34.1
12 9451 7.81 46,807 38.7
based on the literature review.
13 6239 6.3 29,961 30.3
The behavior of the specific energy and total specific energy has 14 4603 4.65 23,052 23.3
been studied in order to analyze the effect of the process parame- 15 7753 6.41 31,499 26.0
ters on the cutting efficiency. The results has been modelled using 16 5468 4.52 23,910 19.8
the RSM non linear model and the coefficient of the regression 17 6960 6.96 39,790 39.8
18 4350 4.35 24,583 24.6
Formula (3) are reported, respectively, in Table 4 and Table 5.
19 4692 5.86 29,517 36.9
A graphical representation of these formulas are reported in 20 6404 5.34 31,226 26.0
Figs. 6 and 7 respectively. 21 6494 6.49 41,502 41.5
The most evident trend is in the analysis of the total specific 22 4981 4.98 24,985 25.0
23 4332 5.68 28,004 35.0
energy, where, due to the high impact of the auxiliary system on
24 5834 4.86 29,695 24.7
the power consumption, the minimum value is always obtained 25 4728 4.73 29,547 29.5
when the material removal rate is maximized. The reason is that 26 5062 5.06 29,876 29.9
the auxiliary systems have a time dependent power consumption 27 4900 4.9 29,710 29.7
and the higher material removal rate accounts for the reduced time 28 5269 5.27 30,123 30.1
29 5122 5.12 29,926 29.9
needed for the operation. Given this first result for the total specific
30 5215 5.22 30,048 30.0
energy, the best option is to maximize the process parameters. This 31 5701 5.7 30,542 30.5
trend could be limited by the effect of tool wear but, under the
314 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316

Table 4 control strategies and more efficient auxiliary systems (i.e., a

Coefficients of the equation for specific energy. different management of the compressed air use). When the frac-
Term Coefficients tion of energy usage related to the auxiliary systems is decreased,
c (J/mm3) 110.7
the importance of the specific energy due to the motors directly
vc (m/min) 1.074 involved in the cutting process (spindle and axis) will acquire
ae (mm) 66.76 higher relevance. The study of the behavior of the specific energy
ft (mm/tooth) 48.51 shows that the optimal condition is attained not maximising the
ap (mm) 5.299
material removal, but the analysis of the function presents a local
vc (m/min)$vc (m/min) 0.002450
ae (mm)$ae (mm) 23.12 minimum inside the range of process parameters analyzed. This is
ft (mm/tooth)$ft (mm/tooth) 1177 an useful information for the design of new, more efficient machine
ap (mm)$ap (mm) 0.1487 tools. In particular the effects of the radial engagement (ae) and feed
vc (m/min)$ae (mm) 0.3118
per tooth (ft) present a different effect on specific energy with
vc (m/min)$ft (mm/tooth) 0.6375
vc (m/min)$ap (mm) 0.02243
respect to the total specific energy. The optimal values in these
ae (mm)$ft (mm/tooth) 227.9 cases are intermediate. The optimal value of the radial engagement
ae (mm)$ap (mm) 1.868 to minimize the specific energy related only to the efficiency of the
ft (mm/tooth)$ap (mm) 14.12 cutting is achieved by maintaining the value suggested by the tool
manufacturer, about 1 mm; the feed per tooth shows an optimal
design value for a 0.12 mm/tooth, a value that is larger than that
Table 5 suggested by the manufacturer but not as high as the upper value
Coefficients of the equation for total specific energy.
tested during this activity.
Term Coefficients

c (J/mm3) 512.7
vc (m/min) 4.779 5. Conclusions
ae (mm) 272.8
ft (mm/tooth) 231.8
The proposed model for the NMV1500DCG milling machine
ap (mm) 18.83
vc (m/min)$vc (m/min) 0.005736 highlights some characteristic behaviors for the power consump-
ae (mm)$ae (mm) 38.61 tion during a machining process. The first important result is that
ft (mm/tooth)$ft (mm/tooth) 3716 the idle or basic state constitutes the larger component for the
ap (mm)$ap (mm) 0.01383 power consumption of the machine; this result is also demon-
vc (m/min)$ae (mm) 1.831
vc (m/min)$ft (mm/tooth) 3.458
strated by many other papers in the literature. This characteristic of
vc (m/min)$ap (mm) 0.1318 the machine could be used by machine tool manufacturers to
ae (mm)$ft (mm/tooth) 745.8 design more efficient machining processes; first steps could be the
ae (mm)$ap (mm) 10.31 reduction of the time during which the machine stays in the ready
ft (mm/tooth)$ap (mm) 35.41
state, the reduction of the moving mass of the machine, and the
introduction of more environmentally friendly lubrication pro-
cesses such as MQL or dry machining.
machining conditions proposed by the manufacturer, this problem The model developed highlights that to obtain a lower envi-
is not critical. Although the final user of the machine tool may be ronmental footprint it is necessary to increase the MRR as far as
interested in the optimization of the total energy consumption of possible by choosing a cutting speed, feed rate, and chip section
the process, the analysis of the specific energy related only to the that are as large as possible while remaining compatible with the
cutting process is also interesting. The reason for this interest is feasible working parameters of the tool. However if only the cutting
given by the actual trend of the Standards and machine tool man- energy is studied, the need to find a local optimum that is not
ufacturers, which are going to introduce strategies to reduce the obtained by the maximization of the process parameters arises.
impact of auxiliary system consumption by developing newer This behavior of the cutting energy will become crucial when the

Fig. 6. Effect of process parameters on specific energy (J/mm3).

 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316
Fig. 7. Effect of process parameters on total specific energy (J/mm3).
power consumption of the support systems is reduced and com-
prises a smaller fraction of the total energy.
Further work is ongoing to extend the model in order to include
coordinate movements such as a general toolpath or to add also
coordinate axis movements generated using interpolated move-
ments using ISO commands such as G02 and G03.
Acknowledgements
The authors wish to thank the Machine Tool Technology
Research Foundation (MTTRF) for providing the machine tool for
the tests and for the general support of their research activities.
Special thanks are reserved for the staff of Mori Seiki for all the
support provided during the installation and use of the machine.
References
Anastas, P., Zimmerman, J., 2004. Design through the twelve principles of green
engineering. Environ. Sci. Technol. 37, 94e101.
Balogun, V.A., Mativenga, P.T., 2013. Modelling of direct energy requirements in
mechanical machining processes. J. Clean. Prod. 41, 179e186.
Bhushan, R.K., 2013. Optimization of cutting parameters for minimizing power
consumption and maximizing tool life during machining of Al alloy SiC particle
composites. J. Clean. Prod. 39, 242e254.
Campatelli, G., 2009. Analysis of the environmental impact for a turning operation
of AISI 1040 steel. In: Proc. IPROMS 2009 Virtual Conference.
Campatelli, G., 2013. Reducing the environmental footprint of machining opera-
tions. In: Proc. HSS Forum 2013 Conference, Florence.
Caspersen, N.I., Sorensen, A., 1998. Improvements of products by means of lifecycle
assessment: high pressure cleaners. J. Clean. Prod. 6 (3e4), 371e380.
Cetin, M.H., Ozcelik, B., Kuram, E., Demirbas, E., 2011. Evaluation of vegetable based
cutting fluids with extreme pressure and cutting parameters in turning of AISI
304L by Taguchi method. J. Clean. Prod. 19 (17e18), 2049e2056.
Dahmus, J., Gutowski, T., 2004. An environmental analysis of machining. In: Proc.
ASME Inter. Mech. Eng. Congr. R&D Expos, Anaheim, California USA.
Devillez, A., Schneider, F., Dominiak, S., 2007. Cutting forces and wear in dry
machining of Inconel 718 with coated carbide tools. WEAR 262 (7e8), 931e942.
Devoldere, T., Dewulf, W., Deprez, W., Willems, B., Duflou, J., 2007. Improvement
potential for energy consumption in discrete part production machines. In:
Takata, S., Umeda, Y. (Eds.), Advances in Life Cycle Engineering for Sustainable
Manufacturing Businesses. Springer, London, pp. 311e316.
Dhar, N.R., Ahmed, M.T., Islam, S., 2007. An experimental investigation on effect of
minimum quantity lubrication in machining AISI 1040 steel. Int. J. Mach. Tools
Manuf. 47, 748e753.
Diaz, N., Helu, M., Jarvis, A., Tonissen, S., Dornfeld, D., Schlosser, R., 2009. Strategies
for Minimum Energy Operation for Precision Machining. Proc. of MTTRF 2009
Annual Meeting PRC, Shanghai.
Diaz, N., Redelsheimer, E., Dornfeld, D., 2011. Energy consumption characterization
and reduction strategies for milling machine tool use. In: Geoloc. Sol. for Sust. in
315
Manuf. e Proc. of the 18th CIRP Int. Conf. on Life Cycle Eng., Braunschweig,
pp. 263e267.
Draganescu, F., Gheorghe, M., Doicin, C.V., 2003. Models of machine tool efficiency
and specific consumed energy. J. Mat. Proc. Tech. 141 (1), 9e15.
EPTA, 2007. Study for Preparing the First Working Plan of the Eco-design Directive
e Report for tender No.: ENTR/06/026.ec.europa.eu/enterprise/policies/sus-
tainable-business/files/workingplan_finalreport_en.pdf. Europe’s Energy Portal,
2012 www.energy.eu.
Fnides, B., Yallese, M.A., Mabrouki, T., Rigal, J.F., 2011. Application of response sur-
face methodology for determining cutting force model in turning hardened AISI
H11 hot work tool steel. Sadhana 36 (1), 109e123.
Fratila, D., Caizar, C., 2011. Application of Taguchi method to selection of optimal
lubrication and cutting conditions in face milling of AlMg3. J. Clean. Prod. 19 (6e
7), 640e645.
Gutowski, T., Dahmus, J., Thiriez, A., 2006. Electrical energy requirements for a
manufacturing process. In: Proceedings of 13th CIRP Int. Conf. on Life Cycle
Eng., Leuven.
He, Y., Liu, F., Wu, T., Zhong, F.P., Peng, B., 2012. Analysis and estimation of energy
consumption for numerical control machining. Proc. Inst. Mech. Eng. Part B J.
Eng. Manuf. 226 (B2), 255e266.
Hicks, C.R., 1993. Fundamental Concepts in the Design of Experiments, fourth ed.
Saunders College Publishing, Holt.
Hill, W.J., Hunter, W.G., 1966. A review of response surface methodology: a litera-
ture survey. Technometrics 8, 571e590.
Horng, J.T., Liu, N.M., Chiang, K.T., 2008. Investigating the machinability evaluation
of Hadfield steel in the hard turning with Al2O3/TiC mixed ceramic tool based
on the response surface methodology. J. Mat. Proc. Tech. 2008 (3), 532e541.
Kellens, K., Dewulf, W., Overcash, M., Hauschild, M.Z., Duflou, J.R., 2012. Method-
ology for systematic analysis and improvement of manufacturing unit process
life cycle inventory (UPLCI) CO2PE! initiative (cooperative effort on process
emissions in manufacturing), Part 2: case studies. Int. J. Life Cycle Ass. 17 (2),
242e251.
Kordonowy, D.N., 2001. A Power Assessment of Machining Tools. Massachusetts
Institute of Technology, Cambridge, MA, USA. B.S. thesis, Department of Mech.
Eng.
Kuram, E., Ozcelik, B., Bayramoglu, M., Demirbas, E., Simsek, B.T., 2013. Optimization
of cutting fluids and cutting parameters during end milling by using D-optimal
design of experiments. J.Clean.Prod. 42, 159e166.
Lawal, S.A., Choudhury, I.A., Nukman, Y., 2013. A critical assessment of lubrication
techniques in machining processes: a case for minimum quantity lubrication
using vegetable oil-based lubricant. J. Clean. Prod. 41, 210e221.
Li, W., Kara, S., 2011. An empirical model for predicting energy consumption of
manufacturing processes: a case of turning process. Proc. Inst. Mech. Eng. Part B
J. Eng. Manuf. 225 (B9), 1636e1646.
Mead, R., Pike, D.J., 1975. A review of response surface methodology from a bio-
metric viewpoint. Biometrics 31, 803e851.
Montgomery, D.C., 2001. Design, Analysis of Experiments, fifth ed. JohnWiley &
Sons, New York, USA.
Mori, M., Fujishima, M., Inamasu, Y., Oda, Y., 2011. A study on energy efficiency
improvement for machine tools. CIRP Ann. e Manuf. Tech. 60, 145e148.
Neugebauer, R., Drossel, W., Wertheim, R., Hochmuth, C., Dix, M., 2012. Resource
and energy efficiency in machining using high-performance and hybrid pro-
cesses. In: 5th CIRP Conf. on High Performance Cutting, Zurich.
Noordin, M.Y., Venkatesh, V.C., Sharif, S., Elting, S., Abdullah, A., 2004. Applica-
tion of response surface methodology in describing the performance of
316 G. Campatelli et al. / Journal of Cleaner Production 66 (2014) 309e316

coated carbide tools when turning AISI 1045 steel. J. Mat. Proc. Tech. 145, Zackrisson, M., 2005. Environmental aspects when manufacturing products mainly
46e58. out of metals and/or polymers. J. Clean. Prod. 13 (1), 43e49.
Orecchini, F., 2000. The ISO 14001 certification of a machine-process. J. Clean. Prod.
8 (1), 61e68.
Ostaeyen, J.V., 2010. CO2PE! (Cooperative Effort on Process Emissions in
Glossary
Manufacturing) e Taxonomy. www.mech.kuleuven.be/co2pe!/taxonomy.php
(accessed January 2012). P1: power need in basic machine state W
Pusavec, F., Krajnik, P., Kopac, J., 2010a. Transitioning to sustainable production P2: power need in idle machine state W
e Part I: application on machining technologies. J. Clean. Prod. 18 (2), 174e P3: power need in cutting machine state W
184. T1: timespans of basic machine state activation s
Pusavec, F., Kramar, D., Krajnik, P., Kopac, J., 2010b. Transitioning to sustainable T2: timespans of idle machine state activation s
production e part II: evaluation of sustainable machining technologies. J. Clean. T3: timespans of cutting machine state activation s
Prod. 18 (12), 1211e1221. Edir: total direct energy requirement J
Rajemi, M.F., Mativenga, P.T., Aramcharoen, A., 2010. Sustainable machining: se- ecut: specific cutting energy J/mm3
lection of optimum turning conditions based on minimum energy consider- k: constant instantaneous specific cutting energy J/s
ations. J. Clean. Prod. 18 (10e11), 1059e1065. b: constant specific cutting energy due to the machine power use in basic state
Shashidhara, Y.M., Jayaram, S.R., 2010. Vegetable oils as a potential cutting fluid e J/mm3
An evolution. Trib. Int. 43 (5e6), 1073e1081. MRR: Material Removal Rate mm3/s
Sokovic, M., Mijanovic, K., 2001. Ecological aspects of the cutting fluids and its in- E: energy needed for the machining operation by the spindle and axis only J
fluence on quantifiable parameters of the cutting processes. J. Mat. Proc. Tech. Es: specific energy needed for the machining operation by the spindle and axis only
109 (1e2), 181e189. in order to remove 1 mm3 of material J/mm3
TERNA, 2012, www.terna.it. E tot: total energy considering all the consumption sources of the machine J
United Nations Framework Convention on Climate Change, 1997. Kyoto Protocol, Es tot: total specific energy considering all the consumption sources of the machine
Kyoto. in order to remove 1 mm3 of material J/mm3
Vijayaraghavan, A, Dornfeld, D, 2010. Automated energy monitoring of machine vc: cutting speed m/min
tools. CIRP Ann.- Manuf. Tech. 59 (1), 21e24. ae: radial engagement mm
Weinert, K., Inasaki, I., Sutherland, J.W., Wakabayashi, T., 2004. Dry machining and ft: feed per tooth mm/tooth
minimum quantity lubrication. CIRP Ann. e Manuf. Tech. 53 (2), 511e537. ap: depth of cut mm