Renewable and Sustainable Energy Reviews 139 (2021) 110738

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: http://www.elsevier.com/locate/rser

Development of process performance simulator (PPS) and parametric

optimization for sustainable machining considering carbon emission, cost
and energy aspects
A.M. Khan a, c, L. Liang a, M. Mia b, M.K. Gupta c, Z. Wei a, M. Jamil a, *, H. Ning a, **
a
College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China
b
Department of Mechanical Engineering, Imperial College London, SW7 2AZ, London, United Kingdom
c
Key Laboratory of High Efficiency and Clean Mechanical Manufacture, Ministry of Education, School of Mechanical Engineering, Shandong University, Jinan, PR China

A R T I C L E I N F O A B S T R A C T

Keywords: The manufacturing industries consume one-third of global energy. Intensive use of electrical energy in industries
Sustainable manufacturing provides the researchers and experts strong reasons to develop and propose strategies to minimize energy
Al2O3 nanofluid consumption. Evidently, the advanced technologies such as nanofluid minimum quantity lubrication in
Energy consumption
machining and process optimization accompanied by holistic models can eliminate the use of conventional fluids
3E
Carbon emission
to reduce production cost and to cope up with environmental issues of global warming and climate change. This
Minimum quantity lubrication study investigates the holistic analysis of four metrics, i.e., surface quality, energy, cost, and carbon emission that
influence the impact of the machining process on the environment of China. Al2O3 based nanofluid was prepared
and used in the external turning of Haynes 25 alloys to improve the machining and to promote sustainability.
Multi-objective optimization was performed to find out a trade-off relation for product quality, energy con­
sumption and production cost. Results showed that the minimum levels of energy consumption and carbon
emission were obtained at the high levels of feed rate and cutting speed. The Carbon Emission Factors (CEF) of
used resources have the most significant effects on CO2 emissions. Furthermore, the feed rate was found to be the
most significant parameter on the machining performance indices. The application of nanoparticles helped to
reduce the cutting energy and CO2 emission, which are proportional to electricity consumption. A holistic
component activity-based cost model was developed, and it was noted that the overhead and workpiece cost
shared more than 95% of the total cost. Multi-objective optimization reduced specific energy by 18.10%, carbon
emission by 22.17% and product cost by 16.25%. Moreover, the present study deals with 3E, i.e., Energy,
Environment, and Economy. The optimum cutting parameters obtained from the concept of 3E at the machine
shop level can significantly improve efficiency of nanofluid MQL assisted machining (NFMQL) process, reduce
cost per unit of product, and achieve the low carbon manufacturing goal.

sustainable, environmentally benign and energy-efficient methods for

production and manufacturing. Innovative strategies and production
1. Introduction
processes are desired to curtail resource consumption and, ultimately,
save energy.
Rapid rise in human population and its growing demand for energy,
The collective wisdom considers the manufacturing sector to be the
over the last century can easily be identified as the key cause of ever-
backbone of an economy. It’s contribution to the net gross domestic
increasing consumption of available natural resources including food,
product is significant for a production-based economy. In the traditional
water, and energy. Excessive use of fossil fuel-based energy resources
manufacturing process, ‘Machining’ is ubiquitous. Less efficient as well
has acted as a catalyst in exacerbating climate change. However, from
as old machine tools, and hoary methods of machining are often
consumer’s perspective, depletion of fossil fuel reserve is also a source of
responsible for unnecessary energy consumption and are sources of
concern. The onus is on the researchers to develop alternative but

* Corresponding author.
** Corresponding author.
E-mail addresses: [email protected] (A.M. Khan), [email protected] (L. Liang), [email protected] (M. Mia), [email protected]
(M.K. Gupta), [email protected] (M. Jamil), [email protected] (H. Ning).

https://doi.org/10.1016/j.rser.2021.110738
Received 5 December 2019; Received in revised form 7 October 2020; Accepted 13 January 2021
Available online 21 January 2021
1364-0321/© 2021 Elsevier Ltd. All rights reserved.
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Abbreviation Q(F) Cutting fluid flow rate

Q(V) Volume of cutting fluid used
3E Energy, Environment, and Economy C(m) Machining cost
ANOVA Analysis of Variance C(i) Idle cost
CAC Component Activity-Based Cost
C(a) Air-cutting cost
CE Carbon Emission
C(tc) Tool change cost
CEF Carbon Emission Factor
C(o) Overhead cost
CF Carbon Factor
C(CT) Cutting tool cost
CMI Chinese Manufacturing Industry
CNC Computer Numerical Control C(Fp) Fluid preparation cost
EAs Evolutionary Algorithms C(Cp) Cost due to cleaning (workpiece)
FA Firefly Algorithm C(chip) Cost of Chips cleaning
GDP Gross Domestic Product P(sb) Standby power
LCM Low Carbon Manufacturing P(su) Setup power
MQL Minimum Quantity Lubrication P(c) Power during cutting
MRR Material Removal Rate P(l) Lubrication power
MRV Material Removal Volume t(sb) Standby time
NFMQL Nanofluid Assisted MQL t(su) Setup time
ONT Ontario t(c) Cutting time
PPC Portable Power Clamp
t(l) Lubrication time
PSO Particle Swarm optimization
E(sb) Standby energy
SDS Sodium Dodecyl Sulphate
E(su) Setup energy
SEC Specific Energy Consumption
RSM Response Surface Methodology E(c) Cutting energy
E(l) Energy consumption for lubrication
Symbols E(d) Disposal energy
P(t) Total power T (L) Tool life
P(i) Idle power CE(P) Carbon emission per part
P(a) Air-cutting power CE(m) Carbon emission during (metal cutting) machining
P(tc) Tool change power f Feed rate
T (t) Total time V (Fp) Volume of prepared fluid
T (Fix) Fixture (clamp) life CF(Fix) Carbon footprints of fixture
t(i) Idle time CF(Np) Carbon footprints of nanoparticles
t(a) Air-cutting time CF(Fp) Carbon footprints of nanofluid preparation
t(tc) Tool change time CF(d) Carbon footprints of disposal activities
E(t) Total energy CE(chip) Carbon emission due to chip
E(i) Idle energy C(indirect) Cost due to indirect activities
E(a) Air-cutting energy C(su) Setup cost
E(tc) Energy consumption for tool change C(c) Cost due to metal cutting
E(cool) Coolant pump energy C(l) Lubrication cost
l(c) Cutting length C(Wp) Workpiece cost
l(a) Air-cutting length C(Fix) Fixture cost
CE(ancillary) Carbon emission due to ancillary activities C(Np) Cost of nanoparticles
V (Wp) Workpiece removal volume C(d) Disposal (fluid) cost
V (d) Disposed volume of cutting-fluid E(cool) Energy consumed by coolant pump
CF(Wp) Carbon footprints of workpiece C(P) Cost per part
CF(CT) Carbon footprints of cutting tools C(env) Environmental cost
CF(F) Carbon footprints of fluid
CF(Cp) Carbon footprints of cleaning (workpiece) process

waste of substantial amount of electricity. In the case of China, more equipment and environmentally unbenign material processing
than 70% of electricity is produced by coal, and the burning of coal has methods with more efficient and ecofriendly material processing
the highest carbon footprints [1]. The demand for energy consumption methods.
in China is gradually increasing, and in 2018, 38% of total electrical Advanced manufacturing industries involve the use of state-of-art
energy was consumed as compared to the energy consumed in 2012. It is technologies such as laser-assisted material processing, electron beam,
important to mention that, in China, in 2015, 83.79% of total energy chemical reactions, micro-manufacturing, and precision manufacturing.
consumed by industrial sector was shared by only manufacturing sector. The integration of sophisticated control systems using Internet of Things
Similarly, metal processing industry consumed 46.6% of total energy (IoT), cloud computing, advanced algorithm and computer integrated
consumed by manufacturing sector [2], thereby intensifying environ­ manufacturing systems greatly augment productivity. However, all
mental issues in future. Metal processing industries can help rein in this these novelties can be labeled sustainable if the production system runs
anticipated rise in the energy demand by replacing less-efficient on improved energy efficiency, use best quality products, cause minimal

2
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

carbon emission, and have low cost of product. Considering the huge serious pollutions to the environment. As such, the manufacturing
impact of machining on the overall manufacturing realm, the ‘industrial sector, particularly the metal processing industry is trying to reduce CO2
personnel’ want their enterprise to be both ‘sustainable’ and ‘profitable’. emissions to avoid the carbon tax and to abate global warming in line
This demands for a more refined approach in designing sophisticated with the global campaign [12]. Due to the high-pressure of new rules
machining systems. For that, cryogenic cooling, high-pressure jet cool­ and regulations implemented by environmental protection agencies
ing, solid lubrication and spray cooling in machining have been evalu­ (such as ISO 14955-1), the Low Carbon Manufacturing (LCM) is
ated by researchers. Perhaps now is the most opportune time to further becoming indispensable for the metal processing industry. The indus­
refine the ‘machining’ using nano-particle assistance in spray based trial sector in China is responsible for around 80% of the total carbon
minimum quantity lubrication that may eventually lead to energy- emission in the country. Thereby, the identification of carbon footprints
efficient manufacturing with low carbon footprints on environment of used electricity and resources are pre-requisites to reduce CO2 to
and less resource consumption. achieve the goal of low carbon manufacturing. To emphasize, Li et al.
It has been learned that it is possible to the machining system to [13] developed an analytical method to estimate the carbon emission
perfection from the perspective of economy, environment and product during the cylindrical turning of AISI 1045, consisting of carbon emis­
quality. With the aim of achieving these objectives, present research sions due to electricity, emission due to the cutting tool, coolant, ma­
constitutes the holistic cost model, carbon emission model and energy terial and chips produced. Nevertheless, the coolant was supplied in a
model within new system boundaries for nanofluid assisted cooling- conventional way, which is not very conducive to achieve the objective
lubrication systems impinged on machining domain. of sustainability. Finally, the study concluded that higher levels of cut­
ting speed guarantee to produce low-carbon manufactured products. Yi
2. Literature review et al. [14] concluded that minimum carbon emission could not always
be achieved at optimal cutting parameters in the external turning pro­
In order to give them adequate weight, a literature review on each of cess. The study emphasized on the multi-objective optimization to find
energy consumption, environmental burden, machining cost, and mul­ the tradeoff solutions. Branker et al. [7] analyzed the CO2 emissions in
tiobjective optimization is presented below. the single point incremental forming of the thin aluminum sheet
considering a holistic system boundary. The authors concluded that the
2.1. Energy consumption workpiece material contributed to the maximum portion of the cost and
CO2 emissions. Tridech and Cheng [15] developed a model for the
Electrical energy consumption characteristics of machine tools are assessment of carbon emission at shop floor model. Lower carbon di­
complex and vary with different stages of the machining process and oxide emissions from used resources, energy efficiency, waste minimi­
functionality state of a machine tool. Dahmus and Gutowski [3] zation, and resource utilization were considered as response to achieve
explored the energy efficiency of machine tools and investigated the sustainable process. It is necessary to develop a correlation between the
environmental impacts of machine tools under narrow system bound­ electrical energy consumption of local power grid and CO2 emissions.
aries. Santos et al. [4] estimated the energy consumption of press-brake; For this reason, Jesweit et al. developed a correlation between electrical
Mori et al. [5] analyzed energy consumption in the drilling process; energy consumption and CO2 emissions [16]. In the present study, same
Davim [6] optimized cutting parameters in external turning process to correlation method will be used to find the Carbon Emission Factor
minimize electrical energy consumption – all these studies considered (CEF).
power consumption to study machine tool efficiency. Yan and Li [7]
concluded that energy and power consumption have a different trend 2.3. Machining cost
with increasing values of Material Removal Rate (MRR) in mechanical
machining. Further, power-based process optimization cannot help to The cost modelling for manufacturing and/or mechanical machining
reduce energy consumption. Later, although the researchers started processes can be divided into two categories – micro-level cost model­
focusing on energy consumption [7,8]; most of them mainly focused on ling and macro-level cost modelling. Macro-level cost models are often
cutting energy without taking into account the energy consumed in employed on batches of discrete products whereas micro-level cost
other machining stages. models are focused on the stage-level of the process. The cutting speed
Recently, the researchers focused on energy consumption at different and tool feed are mostly optimized to get the minimum cost of the
functionality states of a machine tool to lower its total energy con­ products [17–19]. Cutting velocity was reported to have the most
sumption. Li et al. [9] developed models for energy consumption in the influencing effect on process cost [7]. Tipnis [20] published a review of
Computer Numerical Control (CNC) machining considering energy the conventional cost models; thus, the fundamental of cost modeling
consumption at setup, air cutting, cutting and tool change stages. This can be studied from the cited reference. Reportedly, smaller system
study also addressed the cutting time with varying features of a work­ boundaries were considered for the cost assessment of the process with
piece. However, the experiments were performed in dry cutting envi­ the changing environment. The technology-assisted machining pro­
ronment due to which the results are not applicable to other machining cesses are becoming more efficient and sustainable. Branker et al. [12]
environments with coolants or lubricants. Zhou et al. [10] optimized are considered the first authors to add carbon costs in the economic
cutting parameters, measured carbon emission and calculated model during the forming process. Formerly, researchers mostly
machining cost at different stages of turning process. The findings of this considered traditional components to develop cost models [21]; how­
study are also restricted to dry conditions and cannot be generalized. ever, recently nontraditional components are noted with significant ef­
Yoon et al. [11] published a state of the art review study wherein authors fect on the total cost [22].
discussed many energy-saving strategies. The authors envisaged that Machining with nanofluid assisted lubrication technique is an
advanced cutting technologies would better serve the purpose of saving emerging topic from the eco-perspective and sustainability [23–25].
energy consumption. Overall, analysis of literature allows identifying Recently, Padmini et al. [26] calculated the nanofluid cost in the
that the modelling and optimization of energy consumption in the machining process. The authors concluded that nanofluids are very
nanofluid assisted turning process are missing and needed to be useful in the machining process and are the best candidate for the
addressed. replacement of conventional cooling methods. Laurent et al. [27]
concluded that the cost of CO2 is a critical factor and carbon footprints
2.2. Environmental burden can be used as a sustainability criterion in the manufacturing processes.
Hence, the addition of new resources (components) and new function­
The carbon emission due to the manufacturing processes, causes ality stages of machine tools (stages) in the cost modelling requires a

3
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

new design of the process and new optimization of the cutting param­ 3. Mathematical modeling
eters. In summary, the holistic cost modelling of the nanofluid-assisted
machining process including electrical energy consumption cost (en­ The mathematical models for each metric have been developed
ergy cost) of machine tool, nanoparticles cost due to nanofluid prepa­ based on bottom-up-approach. Usually small system boundary is
ration, nanofluid cost, and carbon costs (environmental cost) are considered to study the energy consumption of the machining process –
missing in the literature. it is not enough to understand the shares of each process activity in the
measured response. To solve this issue, a new system boundary, con­
sisting of machining stages, ancillary activities and components, has
2.4. Optimization of the machining process
been considered.
In general, each metric is a sum of outcomes at different stages of
Single objective optimization deals with one response to get opti­
machining. Eq. (1) is a fundamental equation for each metric. Hence, the
mum cutting parameters. However, this solution is infeasible to be
power consumption during machining stages is shown in Eq. (2).
implemented in industrial use. Hence, a multi-objective optimization is
essential to solve complex problems and to find the trade-off between ∑
n
Metric = stage(i) (1)
conflicting responses. Energy measurement and the development of an i=1
energy consumption model from the measured data are the key steps to
achieve sustainability. Many researchers developed a theoretical models P(t) (t) = P(i) (t) + P(su) (t) + P(a) (t) + P(c) (t) + P(tc) (t) + P(l) (t) (2)
[28], empirical models [29,30], discrete-event based model [31], and
hybrid models [32] to study energy consumption behavior in mechan­ In Eq. (2), P(t) is total power that is a sum of six different compo­
ical machining processes. Initially, theoretical models were considered nents. P(su) is the setup power of the CNC machine tool and is measured
as powerful tools for energy-based analysis. However, due to the as 400 Watts. P(i) , P(a) , P(c) , P(tc) , and P(l) represent idle, air cutting, cut­
advancement of the machining process, theoretical models were difficult ting, tool change and lubrication powers, respectively. Power data for
to obtain. To cope with the complicated situation, the empirical models various functionality states of a machine tool are given in Appendix A.1.
were developed, but empirical models provided the approximation of Similarly, total time T(t) is a sum of times during idle t(i) , setup t(su) ,
the machining process. Moreover, many experiments made empirical air-cutting t(a) , cutting t(c) , tool change t(tc) , and lubrication t(l) . The setup
models system dependent. As such, there is a necessity to implement time is a sum of setup time of cutting tool and workpiece. Components of
heuristic models to perform multi-objective optimization. total time are shown in Eq. (3).
Typically, complex industrial problems are difficult to solve using
T(t) = t(i) + t(su) + t(a) + t(c) + t(tc) + t(l) (3)
statistical and conventional algorithms. Thus, the use of non-
deterministic algorithms becomes inevitable. Evolutionary algorithms Eq. (3) can be further simplified into Eq. (4).
lend us efficient and powerful computational tools in solving complex ( )
l(c)
mathematical system with multiple objective functions and variables – T(t) = t(sb) + 1/f l(a) + 2l(c) + t(tc) × (4)
T(L)
thanks to the advancement of the data-driven modern computation
science. Under the umbrella of non-deterministic algorithms, evolu­ T(t) is the total time and t(sb) is standby time of machine tool – the sum
tionary algorithms (EAs) show better performance to solve non- of idle time and setup time (t(i) + t(su) ). Similarly, l(a) and l(c) are air
differentiable problems. Particle swarm optimization (PSO) [33] and cutting length and actual cutting lengths, respectively.
Firefly algorithm (FA) [34] have been used successfully to solve the Most of the studies used Specific Energy Consumption (SEC) as an
real-time complex problems – these algorithms are developed by evaluating criterion to study the efficiency of a machining process [18].
inspired from the behavior of certain insects. The assignment of unequal Here, the electrical energy consumption of a machine tool is studied
weights to each response is also very important consideration to obtain during six various stages of machining process (functionality states of
suitable fitness function for optimal solution [35]. machine tool). The sum-up of all the components of energy is shown in
Eq. (5).
2.5. Aims of the study t(sb)
∫ t(a)
∫ t(c)
∫ t(tc)
∫
Et = P(sb) (t)dt + P(a) (t)dt + (P(c) (t) + P(l) (t))dt + P(sb) (t)dt (5)
Though work on energy, CO2 emission and cost analysis during the 0 0 0 0
other manufacturing process has been reported under traditional metal
forming process, such a work has not been performed in the nanofluid In Eq. (5), P(sb) is a constant amount of power consumed by the
assisted mechanical machining as far as the authors’ knowledge is machine tool during the standby time and tool change time, and its value
concerned. Further, development of holistic models under new system is given in Appendix 4. SEC can be obtained from Eq. (5), dividing by the
boundaries and the use of robust hybrid evolutionary algorithm (PSO Material Removal Volume (MRV). The volume removed during the
hybrid with (FA) in the analysis and optimization of sustainable machining process can be obtained by the product of time and material
machining in general and nanofluid assisted external turning in partic­ removal rate (MRR) as shown in Eq. (6).
ular are scarce. Development of holistic models under new system MRV = vc × ap × f × tc (6)
boundaries and nanofluid assisted external turning are scarce. The
present study is an attempt in these directions and contains the following vc , ap and f are the selected machining parameters and their numeric
new contributions: (1) development of a holistic models of product cost values and definitions are given in nomenclature table. Therefore, the
in sustainable turning process in a new system boundary (2) analyzing SEC of the machine tool can be expressed in Eq. (7) [18].
the impact of Al2O3 nanofluid assisted machining technology on the ∫ t(sb) ∫t ∫t ∫t ∫t
P(sb) (t)dt+ 0(a) P(a) (t)dt+ 0(c) P(c) (t)dt+ 0(c) P(l) (t)dt+ 0(tc) P(sb) (t)dt
economy and environment of China (3) the proposed solution for least SEC= 0
MRV
emission and resource utilization with the development of performance
(7)
simulator of sustainable machining. The results obtained from the pre­
sent study could assist the metal cutting industry in improving work­ Here, the carbon emission during the machining process is consid­
shops for an energy-efficient machining processes, low cost, and Low ered as shown in Eq. (8). The amount of carbon emission during each
Carbon Manufactured (LCM) products. The implementation of the study functionality state of machine tool has been calculated. The Carbon
in the industry can reduce pollution and drive innovation, minimize Emission Factor (CEF) of power grid was calculated and its value is given
carbon footprints and increase productivity. in Appendix A.4.

4
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Table 1
Chemical composition of the subject material Haynes 25.
Element C Co Cr Fe Ni P S Si W

% Weight 0.1 Balance 20.28 1.58 10.20 <0.005 <0.0008 0.01 14.68

zero. However, V(d) has specific values in Flood assisted machining

Table 2
which depends upon flow rate and application time of cutting fluid.
Specification of CNC lathe BOOHI K50P.
(
Parameters Units Values t(c)
C(P) = E(t) × x(e) + x(Wp) × V(Wp) + x(CT) × + x(F) × Q(F) × t(c)
T(L)
Spindle speed (Min.-Max.) Rev./min 160–1200
Feed velocity range (Min.-Max.) mm/min 1–8000 + x(Np) × V(Np) + x(Fp) × V(Fp) + x(F) × Q(F) × t(c) + x(Cp) × V(Cp) + x(d) × V(d)
Maximum machine tool power kW 7.5 t(c)
Maximum cutting force N 10,615 + x(fix) × + C(o) + C(env) (9a)
T(fix)

C(P) = C(i) + C(su) + C(a) + C(c) + C(tc) + C(l) + C(Wp) + C(CT) + C(Fix) + C(Fp)
Table 3
+ C(F) + C(Np) + C(Cp) + C(d) + C(o) + C(env)
Cutting conditions for turning Haynes-L605 alloy.
(9b)
Parameters symbol Units Level Level Level
1 2 3 The overall cost of sixteen items was considered in the model Eq.
Concentration of nanoparticles np vol. % 0.20 0.70 1.20 (9b). It is worth mentioning that definition of important variables is also
in nanofluid given in nomenclature table (See nomenclature Table). Cost of some
Cutting speed vc m/min 30 45 60 machining stages and consumed resources are given in Appendix A.3.
Feed rate f mm/ 0.08 0.12 0.16 Similarly, Definitions of constants used in mathematical modelling of
rev
each metric are shown in nomenclature Table.

Carbon emissions due to electrical energy consumption are minimal 4. Experimental procedure
and thus, it is necessary to study the CO2 emissions at product level [38].
To demonstrate, a new system holistic system boundary was considered This section briefly elaborates about the experimental design, cutting
for carbon emission at product level. Carbon emission per part CE(P) is conditions, nanofluid based minimum quantity lubrication, and
shown in the Eq. (8a) and it includes carbon emission due to total response measurements.
electrical energy consumed, workpiece, cutting tool, fixture, cutting
fluid, nanoparticle, nanofluid preparation, cleaning and disposal activ­ 4.1. Materials and method
ities. Definitions of variables used in Eqs. (8a) and (8b) are given in
nomenclature table. Carbon emissions due to various functionality states Some of the superior characteristics i.e. excellent heat resistance,
of a machine tool are given in Appendix A.2. wear resistance, corrosion resistance, excellent resistance to sulfidation,
low thermal conductivity, and high strength have made Haynes 25 the
CE(P) = CEF × E(t) + CF(Wp) × V(Wp) + CF(CT) × t(c)/T + CF(fix) × W(Fix) best candidate to be used in the biomedical, military and aerospace
(L)

+ CF(F) × Q(F) × t(c) + CF(Np) × V(Np) + CF(Fp) × V(Fp) + CF(Cp) × V(Cp) parts, turbine and furnace parts, and gas turbine engine heat treating
equipment [36]. It is a newly developed cobalt-based superalloy – also
+ CF(d) × V(d)
known as L605. Owing to relatively less thermal conductivity (about 10
(8a) W/m ◦ C), it is considered difficult-to-cut material that causes excessive
tool wear, poor surface finish, difficulties in chip formation, and higher
CE(P) = CE(i) + CE(su) + CE(a) + CE(c) + CE(tc) + CE(l) + CE(Wp) + CE(CT) cutting forces during the machining process. The chemical composition
+ CE(Fix) + CE(Np) + CE(Fp) + CE(Cp) + CE(d) of the workpiece material is shown in Table 1.
(8b) Uncoated carbide inserts (TG-8), used in the present study, were
procured from Zhuzhou Diamond company. The cutting tool was chosen
In Eq. (8b) CE(Wp) represent the carbon emission due to used material.
after doing preliminary trial runs. All experiments were performed on
Similarly, other subscripts represent the carbon emission by the corre­ the CNC turning lathe (model BOOHI SK50P); its specification is listed in
sponding machining stage and component. Table 2.
Component-activity based model was developed for product cost The Blaser vegetable oil was selected due to its excellent results in the
estimation. Total cost during the production of a part is the sum of preliminary test conducted for Haynes 25. The non-toxic and biode­
electrical energy consumption cost, workpiece cost, cutting tool cost, gradable properties of selected oil make the machining process eco-
fixture cost, nanoparticle cost, fluid preparation cost, fluid cost, cost of friendly.
cleaning process, fluid disposal cost, overhead cost and environmental The nanoparticles concentration (vol. %), cutting speed (m/min), and
cost. feed rate (mm/rev) were machining parameters (experimental inde­
In Eq. (9a), x(e) is the cost of electrical energy consumption and its pendent variables) for the present study. Moreover, the levels of
constant value is given in Appendix 4. Similarly, x(Wp) , x(CT) , x(F) , x(Np) , machining parameters were chosen after extensive literature review and
x(Cp) , x(Fp) , x(d) , x(fix), C(env) , C(o) are costs per unit amount of resources initial tests. However, values of depth of cut (1.0 mm) and diameter
used including workpiece, cutting tool, fluid, nanoparticle, cleaning workpiece (50 mm) were kept constant in all experiments. Other pa­
process, fluid preparation, disposal, fixture, environmental and over­ rameters related to MQL technology including nozzle design, nozzle
head, respectively. Similarly, V(Wp) is a volume of material removed; distance from cutting zone and flow rate were kept as constant for all
V(Np) is volume of nanoparticles used, V(Fp) is volume of prepared experiments. These constant parameters were chosen from literature
nanofluid, and V(d) is volume of disposed cutting fluid. In NFMQL review and doing preliminary experiments. The machining parameters
assisted machining, due to minimal amount of use V(d) was considered along with their levels are shown in Table 3. With three parameters and
three levels, 9 experiments were conducted according to standard L9

5
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Fig. 1. Experimental setup (a) Al2O3 nanofluid assisted turning process (b) Power measurement.

HDMI connector, Micro SD Card Moreover, carbon emissions are

Table 4
calculated according to the standard method proposed by Ref. [16]. The
Experimental results for Al2O3 assisted turning process.
costs of consumed resources were collected from industry and from
No. Machining parameters Responses published literature and mentioned in Table 1. Tool life was measured
np vc f Ra (μm) SEC (J/mm3) C(P) ($) CE(m) (g) according to ISO standard, and cutting tools were disposed of when
reached to the maximum criteria of tool wear i.e., 0.3 mm average flank
1 0.2 30 0.08 1.35 117.27 44.60 2.906
2 0.2 45 0.12 1.42 80.18 26.99 1.961 wear or 0.6 mm maximum flank wear. An optical microscope ARTCAM
3 0.2 60 0.16 1.48 52.06 20.87 1.265 (Model: 130-MT-WOM) was used to measure the wear of cutting tools.
4 0.7 30 0.12 1.42 92.85 33.80 2.271 CEF was calculated by the data provided by Nanjing government. The
5 0.7 45 0.16 1.41 61.19 23.56 1.496 practitioners from well-known cutting tool manufacturers (Walter) and
6 0.7 60 0.08 1.29 72.24 28.52 1.755
7 1.2 30 0.16 1.44 74.78 28.53 1.817
automobile industry (Ford Motors) also helped by providing data in­
8 1.2 45 0.08 1.30 82.83 33.80 2.026 ventory for used resources and validated the employed methods.
9 1.2 60 0.12 1.21 57.41 23.55 1.404
5. Results and discussion
Taguchi Array.
The recorded and measured experimental results of surface rough­
The MQL set up along with the CNC lathe machine is shown in Fig. 1.
ness, specific energy consumption, carbon emission, and total cost were
Before starting the experiments, nearly 13% of the total length of the
first analyzed using appropriate statistical indicators and, afterward,
workpiece was fixed into the chuck jaws and tightened properly to avoid
discussed under the perspective of fundamental process physics.
vibration. Each experiment was performed with a different set of
A total of nine experiments were performed by varying 3 sustainable
machining parameters according to the design of the experiment and
parameters i.e. nanoparticles concentration, cutting speed, and feed rate
repeated three times to get the average values for analysis. The nano­
for the NFMQL assisted turning process. Taguchi’s L9 orthogonal array
fluid preparation procedure has been followed from Refs. [24–26]
was employed using MINITAB-18 software. Respective to each run, the
wherein Alumina (Al2O3) was used due to its cost-effective nature [37].
responses were measured and/or computing. These results are shown in
Table 4.
4.2. Response measurement and data collection

Power consumption was measured using Load Controls PPC-3 Power 5.1. Energy consumption
Meter. It was measured during various functionality states (stages) of a
machine tool. The measured power was multiplied with a known period The electrical energy consumption is assumed to be dependent on the
in the cycle time to get electrical energy consumption of various machining parameters, and, therefore, is investigated with respect to the
(functionality states of machine tool) stages of machining process. Smart cutting speed, nano-particle concentration and feed rate. Fig. 3 shows
meter was used to measure the lubrication power. The smart meter 3D plots with the two parameters as inputs (x- and y-axes). The specific
basically consists of Raspberry Pi-3, and Smart-Pi, Keyboard and Mouse, energy consumption is taken as the response (z-axis).

6
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Fig. 2. 3D plots for specific energy consumption with respect to (a) nano-particle concentration and cutting speed; at constant f = 0.12 mm/rev, (b) nano-particle
concentration and feed rate, at vc = 45 m/min (c) cutting speed and feed rate, at np = 0.7 vol %.

Fig. 3. Comparison of coefficient of friction in various cooling methods. Variation of (a) CoF with respect to time, (b) CoF with respect to concentration of
nanoparticles.

Fig. 4. 3D plots for carbon emission with respect to (a) nano-particle concentration and cutting speed; at constant f = 0.12 mm/rev, (b) nano-particle concentration
and feed rate, vc = 45 m/min (c) cutting speed and feed rate, np = 0.7 vol %.

Fig. 3a illustrates that the increase in nanoparticle concentration has nanoparticle concentration increases. This is because an increase in
a minimal effect on the SEC. However, an increase in cutting speed can nanoparticle concentration reduces cutting power consumption. The
significantly reduce the SEC, as also reported by Ref. [43]. Expectedly, plot in Fig. 2 (c) slopes downward with increasing feed rate. This is
the increased tool life due to increased concentration of nanoparticles because the increasing feed rate results in a decrease in cutting time
reduces the tool change time ttc [44]. In Fig. 2b, SEC decreases as the [45]. Therefore, the maximum possible feed rate must be used to get a

7
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

5.2. Carbon emission

Reduction in energy consumption is compulsory to cut down CO2

emissions in the atmosphere [39]. This means that the amount of carbon
emission during a machining process is a function of energy consump­
tion. The relationship between energy consumption and CO2 was stud­
ied by Jeswiet and Kara [47] and the authors developed an innovative
method to calculate carbon emission. The method employed in this work
to calculate carbon emission during the machining stage. The carbon
emission factor of the Nanjing power grid station is 258.35 kg-CO2/MJ
[13].
The influence of machining parameters on CE(m) is portrayed as 3D
plots in Fig. 4. As can be seen, the carbon emission decreases with an
increase in the feed rate and cutting speed. However, the concentration
of nanoparticles has less significant effect on carbon emission. This is
due to a pre-dominant reduction in machining time in the turning pro­
cess [12,48].
Carbon emissions due to electricity consumed by machine tool and
due to other resources and ancillary activities are summed up to get
carbon emission per part. Indicators are important to extract quick in­
Fig. 5. Variation of carbon emission indicator at different cutting conditions. formation from the process. This study develops a new indicator for
carbon emissions. Contribution of electrical energy consumption on the
lower energy consumption of machine tool, as also claimed by Ref. [40]. total CO2 emission per part is called Carbon Emission Indicator (CEI)
The application of nanofluids in cutting process helps to reduce the and can be evaluated as (CE(m)/CE ). It can be seen from Fig. 5 that the
(P)
Coefficient of Friction (CoF) and, consequently, to reduce cutting power contribution of electricity in total carbon emission is minimal. A
[46]. Fig. 3a shows that CoF decreases with the increase in cutting time. maximum of 1.291% CO2 emission occurred in experimental test 4.
Similarly, experiments were performed to analyze the effect of various
concentrations of nanoparticles on CoF. From Fig. 3b, it was noted that,
5.3. Economic analysis
keeping all other machining parameters constant, the parameter of
nanofluids (vol. % = 0.2) reduced CoF by 28.5% and 64.28% as
In this study, a component-activity based cost model has been
compared to conventional (Flood) and dry environment, respectively.
developed specifically for Al2O3 based MQL (NFMQL) assisted turning
However, when the concentration of nanoparticles increases (vol% =
process. The proposed cost model includes cost of six functionality state
1.2), nanofluid produces 124% less CoF as compared to Flood and 187%
of a machine tool, three indirect activities, and seven various resources
less CoF as compared to dry environment, respectively. The results are in
used (as shown in Fig. 6).
good agreement with the findings mentioned in state of art review ar­
The effect of process parameters on the total cost of the product is
ticles [23–25]. The performance of nanofluid is compared with those of
highlighted in Fig. 7. The increase in cutting speed reduces the total cost
the already employed dry and conventional (Flood) methods. It can be
as shown in Fig. 7a. Similarly, it is obvious from Fig. 7b that feed rate
said that nanofluid assisted machining can replace conventional
possesses the most significant effect on the cost and that the minimum
methods.
cost is achieved at the maximum level of feed rate. However, the con­
centration of nanoparticles has the smallest effect on the cost. As dis­
cussed in our previous study [49], the nanoparticles have a significant

Fig. 6. Components and activity-based cost modelling under a new system boundary.

8
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Fig. 7. 3D plots for product cost with respect to (a) nano-particle concentration and cutting speed; at constant f = 0.12 mm/rev, (b) nano-particle concentration and
feed rate, at vc = 45 m/min (c) cutting speed and feed rate, atnp = 0.7 vol %.

Fig. 8. Cost contributions by components and activities in a system boundary. (a) Cost per part, (b) Cost contributions by components, (c) Cost contributions
by activities.

effect on tool wear and tool life. As in the present study, the cutting time
Table 5
was less and a minimal amount of nanoparticles were used, so less effect
Initial learning parameters of PSO and Hybrid PSO-FA.
of nanoparticles has been noted on the total cost. In summary, the
Input parameters Value of parameters minimum cost can be achieved at the maximum possible levels of the
S, number of agent particles 50 cutting parameters without compromising the quality of the product.
Number of iterations 100 Fig. 7c depicts the decreasing trend of C(P) with increases in cutting speed
Maximum permissible inertia weight 1.4
and feed rate.
Minimum permissible inertia weight 0.5
Maximum defined learning rate, C1max = C2max 2 Fig. 8 depicts the total cost incurred by all various machining stages
Minimum defined learning rate, C1min = C2min 1.5 and resources used in machining process within the holistic system
H (constant) 5 boundary. The overhead activity and workpiece contribute the
maximum percentage contribution of 69.03% and 30.06% respectively.
Fig. 8b and c illustrate that overhead and workpiece are the two

9
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

present study, cutting tool cost shares 0.23% of the total cost at the
experimental conditions of test 9. The cost-share difference between two
studies is due to the different system boundaries selected. Klocke and
Eisenblatter [50] did not consider costs of some components and ac­
tivities which authors used in their model. Thus, it can be said that
application of the proposed eco-benign technology can reduce the cost
of cleaning and disposal related activities and improve the economic
pillar of sustainability.

5.4. Multi-objective optimization

The multi-objective optimization of machining parameters used for

turning Haynes 25 alloy is performed by using a constrained hybrid
Particle Swarm Optimization–Firefly algorithm (PSO-FA) – an effective
hybrid algorithm to solve nonlinear multi-objective optimization prob­
lems with multiple variables. The term constrained algorithm has been
used because the input parameters such as nanoparticle concentration,
cutting speed and feed rate were selected in the respective specific
ranges. The responses taken were surface roughness, specific energy
Fig. 9. Convergence characteristics graphs for multi-objective optimization of consumption, carbon emissions and cost of the product. In the end, the
combined objective.
efficiency of the implemented algorithm is confirmed and verified
against the classical PSO and FA methods. The details of PSO can be
found in Ref. [33], and that of the Firefly algorithm can be found on
Table 6
[51]. The complete details for parameters are presented in the following
Control variables and their selected values (for optimal response variables).
sub-steps:
Control variables Combined Values
Step 1. Learning Parameters: To run the algorithms efficiently, the
PSO FA HPSO-FA
initial steps in terms of algorithm learning parameters required to be
% vol. 1.2 1.2 1.2 identified. Though, these parameters decide the performance of the al­
vc 60 60 60
gorithm, still the choice of these parameters is very difficult. In this
F 0.16 0.16 0.16
work, three algorithms, i.e., PSO, FA, and Hybrid PSO-FA were applied
Best Solution 0.09104 0.09101 0.090
Standard deviation 0.830 0.732 0.342 to the selected values. Of these, the Firefly algorithm was the only al­
Average time (sec) 15.03 7.05 3.89 gorithm that does not require any learning parameters. For instance, the
Success rate 75% 80% 95% number of agent particles, number of iterations, maximum permissible
Percentage error 9.43 6.40 2.50 inertia weight and minimum permissible inertia weights were used in
the PSO and hybrid PSO-FA. The values of parameters used in PSO-FA
components that share the maximum contribution; similarly, the are shown in Table 5.
cleaning (workpiece cleaning) stage and cutting stage contribute the Step 2. Generation of the Fitness Function: The next step of this
maximum cost at the metal cutting level (Fig. 8a). optimization model is to generate the fitness function. Initially, the
An experimental study was conducted in the German automotive selected responses are converted into single responses, and then, the
industry by Ref. [50] to estimate the cost per part. It was noted that regression equations for a single response model have been developed
cutting-tool shared 2–4% of total cost of workpiece. According to and employed in the MATLAB program. For combining responses, the

Fig. 10. The primary energy source for a grid.

10
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Table 7
Reduction in CO2 at machining level CE(m) and product level CE(P) .
Exp. No. Time Carbon emissions Primary energy Primary energy
Sec
CE(m) CE(P) Parts produced per day Nanjing grid ONT Grid 100% renewable energy
kg kg No.
CE(m) (kg/Per day) CE(P) (kg/Per day) CE(m) (Per day) CE(m) (Per day) CE(P) (Per day)

1 240.00 0.0029 0.1897 96.00 0.2791 18.211 0.1953 0 17.931

2 105.26 0.0017 0.1653 218.88 0.3811 36.191 0.2667 0 35.810
3 58.82 0.0013 0.145 391.68 0.4956 56.781 0.3469 0 56.286
4 157.89 0.0023 0.1782 145.92 0.3314 25.998 0.2319 0 25.667
5 78.95 0.0015 0.1471 291.84 0.4369 42.918 0.3058 0 42.481
6 117.65 0.0018 0.1781 195.84 0.3438 34.887 0.2406 0 34.544
7 117.65 0.0018 0.1884 195.84 0.3559 36.891 0.2491 0 36.535
8 157.89 0.002 0.1787 145.92 0.2957 26.07 0.2069 0 25.774
9 78.95 0.0014 0.1384 291.84 0.4098 40.38 0.2868 0 39.970

Fig. 11. A comparison of proposed machining technology with commonly used Flood machining.

weighted sum method has been used on the selected response values to to run the program effectively. With this regard, 100 trial runs in terms
generate a single objective. Eq. (10) is used for the combination. of iterations were performed on the combined objective values. The
results from the multi-response optimization are presented in Table 5,
X1 W1 X2 W2 X3 W3 X4 W4 while the convergence characteristics with respect to the combined
Ymin = + + + (10)
X1min X2min X3min X4min objective results are exhibited in Fig. 9. From the optimization results, it
can be stated that the 1.2% concentration volume, 60 m/min cutting
where, X1 , X2 , X3 and X4 represent the surface roughness, specific energy speed and 0.16 mm/rev feed rate are the optimal values that minimize
consumption, carbon emissions and cost of the product. The unequal the combined objective value. Moreover, the surface roughness was
weights W1 = 0.252, W2 = 0.251, W3 = 0.249andW4 = 0.247 were reduced by 12.31%, specific energy by 18.10%, CO2 emission by
assigned to SR, SEC, CE(m) and C(P) , respectively. Determination of 22.17%, product cost by 16.25%. These significant reductions in re­
realistic weight was crucial for more accurate estimation, and for that, sponses are just one initiative towards the goal of greener production.
the Grey Entropy-based weight assignment has been employed. The
detailed weight calculation method can be found in published study The results obtained from the three algorithms were different in
[52]. Initially, Grey Relational Coefficients (GRC) were calculated from which the hybrid PSO-FA shows the minimum values (Table 6).
normalized data and then GRC were used to find weights of each Furthermore, the other results, such as standard deviation, average time,
response. After assigned weights, combination presented in Eq. (10) was success rate, and percentage error were found to be better with the use of
used to obtain a fitness function Y(min) in the MATLAB code of subjected hybrid PSO-FA compared to the employment of PSO and FA separately.
algorithms.
Step 3. Working Procedure: The next task of optimization modelling is

11
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Table 8
Positive impacts of proposed technology on originally used conventional method in industry.
No. Time (s) Q(F) (l/h) Q(V) (litters) E(cool) (J) CE(d) (kg-CO2) CE(chip) (kg-CO2) C(d) ($) C(chip) ($)

Original settings: Flood aided machining

1 240.00 240 0.0066 120000 0.037 0.00611 0.008321 0.0537

2 105.26 240 0.0029 52630 0.016 0.006034 0.00365 0.0530
3 58.82 240 0.0016 29410 0.009 0.005995 0.002039 0.0527
4 157.89 240 0.0043 78945 0.025 0.006034 0.005474 0.0530
5 78.95 240 0.0021 39475 0.012 0.006035 0.002737 0.0530
6 117.65 240 0.0032 58825 0.018 0.005995 0.004079 0.0527
7 117.65 240 0.0032 58825 0.018 0.005995 0.004079 0.0527
8 157.89 240 0.0043 78945 0.025 0.006034 0.005474 0.0530
9 78.95 240 0.0021 39475 0.012 0.006034 0.002737 0.0530

Proposed Technology: NFMQL assisted machining

1 240.00 0.30 0.0087 26315 0 0 0 0

2 105.26 0.30 0.0048 14705 0 0 0 0
3 58.82 0.30 0.0130 39472.5 0 0 0 0
4 157.89 0.30 0.0060 19737.5 0 0 0 0
5 78.95 0.30 0.0097 29412.5 0 0 0 0
6 117.65 0.30 0.0090 29412.5 0 0 0 0
7 117.65 0.30 0.0131 39472.5 0 0 0 0
8 157.89 0.30 0.0065 19737.5 0 0 0 0
9 78.95 0.30 0.0190 60000 0 0 0 0

very high when it is using fossil fuels such as coal, oil and natural gas as
the primary energy source. Electricity produced by using primary energy
sources such as biomass, wind, hydro, wave, solar and geothermal is
known as “Clean Energy” sources, as they own no carbon footprints. To
reduce the carbon emission in the local industry of Nanjing, it is pro­
posed that the local grid must generate Clean Energy (Fig. 10).
Table 7 shows the data for the production of discrete parts in a
Chinese local industry (Nanjing Dongmo Electromechanical
Manufacturing Co., Ltd) per day (8 h shift) with 20% downtime. Nine
parts are produced with varying cutting parameters and production
time. Considering 20% downtime, total parts produced per day were
calculated, and then the carbon emissions per day were estimated. The
results show that if the same experiments were performed in Ontario
(ONT), 70% less CO2 would be emitted to air. This is because the local
grid of Ontario uses renewable resources to generate electric energy
[53]. However, the proposed technology will only work to reduce the
carbon emission during the machining stages. Use of clean energy dur­
ing various machining processes in workshop can only reduce on an
average of 1% CO2 emission per part produced. Hence, it is clear from
Fig. 12. Quantitative holistic comparison of proposed technology with con­ the scenario – using Clean Energy is not enough to reduce environmental
ventional approach. pollution. More efforts are needed to reduce carbon emissions in ancil­
lary activities (Eq. (8b)) involved in part production. The novel findings
force metal processing industry to bring sustainability in ancillary
6. Reduction of resource use, waste, and emissions
components and activities involved in the production of mechanical
parts.
The Manufacturing Industry of China (CMI) is an energy-intensive
Conventional Flood assisted machining processes have been under
but important pillar of the economy. In 2013, the CMI consumed
practice for years. The after-process activities associated with Flood
53.3% of the total electricity demand of the country [1]. Moreover, the
machining make it a non-sustainable process. Efforts have been made to
demand for primary energy consumption is still increasing. Hence,
replace this old expensive cooling method to minimize resource utili­
reducing electrical energy consumption can bring benefits to the entire
zation and carbon emissions. For that, the minimum quantity lubrication
world as the carbon emissions reciprocate the behavior of energy con­
or MQL is a technology that uses a minimum amount of cutting fluids
sumption [41]. Therefore, in the present study, the NFMQL technology
(1000 times less) as compared to conventional cooling methods where
is proposed to reduce resource utilization and carbon emissions.
unbenign emulsions are used. Nanofluid assisted MQL or NFMQL is the
advanced version of MQL technology which provides better perfor­
6.1. Reduction in carbon emission mance than traditional MQL [46].

The current experiments are performed in Nanjing, a city situated in

the south of China [42]. The local grid of Nanjing used more than 70% of 6.2. Eco-benign technology
its electricity by using coal as a basic energy source and possess a very
high CEF of 258.2 kg-CO2/GJ which is 71.3% higher than Ontario (ONT) In comparison to conventional Flood assisted machining, the NFMQL
power grid station and 100% more than the ideal case (Clean Energy) technology is completely different as depicted in Fig. 11. In Flood
[16]. Fig. 10 shows the primary energy sources of a typical power sta­ assisted machining, produced parts and chips need further cleaning,
tion. According to Ref. [53], the carbon emission signature of the grids is drying, and protective methods are adopted to avoid corrosion. Another

12
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Fig. 13. Integration of process performance simulator with the generalized concept of Energy, Environment, and Economy savings as a whole.

machining) and in proposed (NFMQL) assisted machining.

Table 8 reveals that implanting NFMQL technology can reduce 50%
energy consumption as compared to original settings. This can be
explained as the coolant pump motor needs more power to deliver
coolant at higher flow rates. However, for MQL technology, low power-
rated air compressing is enough to deliver air mist into cutting zone, and
consequently consume less power. The percent reduction in energy
consumption could vary according to machining process and machine
tool used. The main disadvantage of current settings is the burden of
after-use activities i.e., machine tool, workpiece, and chips cleaning,
fluid disposal and so on. On the other side, MQL allows eco-benign and
clean machining, it produces dry and clean chips and thus does not
require additional post-machining activities. Consequently, 100% CO2
emission and cost in lieu of fluid disposal and chip cleaning can be
prevented in case of MQL machining.
Additional experiments are performed for the holistic comparison for
the proposed technology with conventional practice. A total of 14 in­
dicators were measured at optimal cutting conditions and compared
with conventional flood aided process. The measured values of each
indicator were first normalized and then plotted. From fourteen in­
dicators, only tool life and productivity are required as maximum the
Fig. 14. Architecture of the performance simulator for sustainable machining.
better. Fig. 12 clearly illustrates that without compromising the net
productivity, proposed Hybrid Al-GnP assisted machining yielded better
problem associated with Flood assisted machining is the disposal of used results for all indicators.
cutting fluid. These additional procedures add cost and make it more
expensive than the NFMQL assisted machining. Fig. 11 also enlightens
6.3. Development of Process Performance Simulator
how authors proposed machining technology cuts the additional steps
(transport, cleaning and disposal related activities) – a key element to
Fig. 13 describes the linkage between total energy consumption, CO2
reduce resource utilization.
emission and total cost. In addition it depicts how carbon emission
In Table 8, Q(F) represents the flow rate of cutting fluid and Q(v) is the
varies with the increase of electrical energy demand and resource-based
volume of cutting fluid used in specific cutting time. The cutting energy
embodied energy consumption [54]. The framework starts with a per­
consumed by the coolant pump is depicted by Ecool and similarly energy
formance simulator, which is a new tool developed to estimate the
consumed by air compressor in MQL technology is depicted by Elub .
metrics and resources consumption.
Carbon emission CE(d) and cost C(d) are due to disposal of cutting fluid.
In the present study, the total energy consumption is a sum of elec­
Similarly, carbon emission CE(chip) and cost C(chip) due to cleaning and
trical energy used by machine tool and embodied energy of consumed
management of chips were calculated in original settings (Flood assisted
resources. The author’s proposed method helps to reduce both types of

13
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Fig. 15. Development of Process Performance Simulator (PPS) for sustainable machining.

energies. On the one side, application of nanoparticles in machining machining method can achieve three pillars of 3E approach i.e., Energy,
process reduced cutting power [46] i.e., less electrical energy con­ Environment and Economy. Furthermore, environmental cost is
sumption. On the other side, the minimal application of fluid not only considered in the production cost, thus, reduction in carbon emission
saves resource consumption, but also removes necessity of the will help to reduce the total production cost. Therefore, the framework
post-activities like cleaning of chips and workpiece and disposal of clearly shows, authors proposed technology can successfully save En­
fluids. Thus, a solution towards less fluid is just one positive outcome. ergy, Environment, and Economy as a whole. In summary, the 3E (En­
Adding to it, an increase in cutting tool life [46], and elimination of ergy, Environment, Economy) concept has been evaluated for a local
cleaning and disposal activities are the significant benefits of proposed Chinese Manufacturing industry.
method. The impact of the employed innovative technology on environment
The use of proposed technology not only minimizes electrical energy and economy is necessary to determine. Thus, a Process Performance
consumption but also reduces resource consumption and thus reduces Simulator (PPS) has been developed with the help of data inventory and
CO2 emissions. The experimental validation of results shown in Fig. 13 proposed empirical models.
can be found in Table 8. Fig. 14 shows the architecture of performance simulator for sus­
At the machining level (bottom level of Fig. 13), energy cost and tainable machining. The architecture consists of three phases i.e., input,
carbon emissions due to machining stages are highly depending upon process, and output. The simulator gets used machining-parameter set­
power consumption. Reduction in power consumption can reduce tings, CNC code, tool path and information about workpiece and ma­
electrical energy consumption and its corresponding footprints. Simi­ chine tool power characteristics from input phase. The inputs are then
larly, energy cost C(m) highly depends on power/energy consumption. processed to estimate energy, cost and carbon emission yielded during
On the top of this framework, amount of consumed resources is the process which stands for the impact on proposed technology in
responsible for their corresponding energy, emissions and cost. Thus, a machining process on environment and economy. For simplicity,
reduction of resources consumption in proposed NFMQL assisted simulator is developed for fewer parameters, the other factors can also

14
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

be added in the simulator to get more precise analysis. • The performed process optimization using robust and innovative
Machining conditions from Exp. no. 9 of Table 4 have been taken as approach - hybrid PSO-FA optimization algorithm - reduced SEC by
an example to verify the simulator. Process, material selection, 18.1%, carbon emission by 22.17%, and product cost by 16.25%.
machining parameters, data inventory, and resource utilization are • Energy-Environment-Economy (3E) integrated methodology was
allocated in the input window (Fig. 15). The proposed holistic empirical proposed for the sustainability assessment of the advanced nanofluid
models process the data to estimate the metrics. The performance assisted machining process. In addition, a performance simulator for
simulator provides an environmental burden due to consumed elec­ sustainable machining was developed to estimate sustainability
tricity and coolant as well as estimates the cost of machining process. metrics of machining process.
Hence, this efficient method can help manufacturing engineers in
effectively analyzing the performance of process. Future recommendation
It is to be noted, the performance simulator is developed to access the
performance of the machining process. Thus, the simulator itself cannot Future research avenues can be the investigation of energy efficiency
save energy consumption (Fig. 15). However, it helps the machinist to of nanofluid assisted machining, correlation of energy consumption and
estimate resource consumption without doing large number of expen­ environmental degradation, assessment of machine tool’s energy con­
sive experiments. A friendly user performance simulator can convert sumption, life cycle assessment of nanofluid MQL assisted machining,
inputs into useful quantitative indicators. Inputs of experimental data and the integrated modeling of above parameters using intelligent al­
were fed into performance simulator to obtain useful information such gorithms. In addition, future research is also focused on to include green
as environmental burden of electricity, environmental burden of cool­ engineering to developed 4E (Energy, Environment, Economy, green
ants, carbon emission per parts, and production cost. Engineering) framework for mechanical machining processes.

7. Conclusion Funding

A system boundary including new resources (components) and The work is supported by the National Key Research and Develop­
various machining stages (activities) has been undertaken to construct ment Project [Grant No. 2018YFB2002205.
novel holistic models for machining time, energy consumption, and
carbon emission to evaluate the overall performance of Al2O3 nanofluid Author contribution statement
in turning experiments. The selection of cutting parameters and their
levels have a significant effect on the machinability and sustainability of Khan A.M: Conceptualization, Methodology, Software, Data cura­
the cutting process. Therefore, the multi-objective optimization was tion, Writing – original draft preparation, Visualization, Investigation,
performed to realize optimum cutting conditions. The following Validation. Liang L: Methodology, Writing – original draft preparation,
important findings are obtained from the study: Supervision, Validation. Mia M: Conceptualization, Methodology,
Writing – original draft preparation, Visualization, Writing- Reviewing
• The nanoparticle concentration in general showed a prominent effect and Editing. Gupta M.K: Software, Visualization, Investigation, Writing-
on the cutting energy. A 1.2% content of nanofluid caused reduction Reviewing and Editing. Wei Z: Methodology, Software, Investigation.
in the cutting power and improvement in the surface finish. Jamil M: Conceptualization, Data curation, Ning, H: Supervision,
• To reduce specific energy consumption, a higher machining feed is Writing- Reviewing and Editing, Funding acquisition
recommended. For effective energy saving, the energy consumption
of various functionality states of machine tool should be reduced. Declaration of competing interest
Nanoparticle assisted machining has more potential for energy
saving, as compared with the traditional one, as cutting fluid pump is The authors declare that they have no known competing financial
absent. interests or personal relationships that could have appeared to influence
• The cost of electrical energy consumption, nanoparticles and nano­ the work reported in this paper.
fluid preparation per part was nearly 1% of the total cost of a part –
making proposed machining technology economical. Acknowledgement
• Carbon emission reciprocated the behavior of energy consumption.
Carbon emission due to electrical energy consumption CE(m) were The paper is one of the jewels of the PhD research work of the first
minimal portion of the total carbon emission per part produced CE(p) . author. The authors want to thank the editor and the anonymous re­
Thus, NFMQL assisted machining process can make process sus­ viewers as their comments helped us improve the quality of the paper.
tainable at the machining level. In summary, proposed method is The authors would also like to acknowledge the local metal processing
sustainable both environmentally and economically. industry in China and efforts made by Prof. I.S. Jawahir from the Uni­
versity of Kentucky (USA) for valuable suggestions.

Appendix A1. Variable components used in power and energy models

No. Pa (W) Pc (W) Pt (W) tc (s) Ea (J) Ec (J) Etc (J) El (J) Et (J)

1 2840 948 3788 240.00 42600 227520 86.878 102000 501806.88

2 2960 1915 4875 105.26 44400 201579 65.789 44737 674360.53
3 3054 2848 5902 58.82 45810 167529 46.136 25000 490032.61
4 3196 1190 4386 157.89 47940 187895 77.973 67105 879649.55
5 2790 2686 5476 78.95 41850 212053 47.133 33553 579765.55
6 2910 1461 4371 117.65 43650 171882 79.761 50000 679965.05
7 2912.31 1662 4574 117.65 43684.65 195529 90.498 50000 703892.79
8 3120 672 3792 157.89 46800 106105 78.75 67105 784720.86
9 2825 2190 5015 78.95 42375 172895 38.04 33553 543886.73

15
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

Appendix A2. Variable components used in empirical model of carbon emission

No. CEa (kg-CO2) CEc (kg-CO2) CEm (kg-CO2) CEWp (kg-CO2) CECT (kg-CO2) CEP (kg-CO2)

1 1.09 × 10 − 4
5.87 × 10− 4
2.90 × 10− 3 0.0911 0.0823 0.1880
2 1.14 × 10− 4
5.20 × 10− 4
1.96 × 10− 3 0.0899 0.0623 0.1653
3 1.18 × 10− 4
4.32 × 10− 4
1.26 × 10− 3 0.0893 0.0437 0.1449
4 1.23 × 10− 4
4.85 × 10− 4
2.27 × 10− 3 0.0899 0.0738 0.1781
5 1.08 × 10− 4
5.47 × 10− 4
1.49 × 10− 3 0.0899 0.0446 0.1470
6 1.12 × 10− 4
4.43 × 10− 4
1.75 × 10− 3 0.0893 0.0755 0.1781
7 1.12 × 10− 4
5.04 × 10− 4
1.81 × 10− 3 0.0893 0.0857 0.1883
8 1.20 × 10− 4
2.73 × 10− 4
2.02 × 10− 3 0.0899 0.0746 0.1786
9 1.09 × 10− 4
4.46 × 10− 4
1.40 × 10− 3 0.0899 0.036 0.1383

Appendix A3. Variable components used in Component-activity based cost (CAC) model

No. Ca ($) Cc ($) Cm ($) CWp ($) CCT ($) Cenv ($) CP ($)

1 1.36 × 10− 3
7.29 × 10 − 3 0.016075 0.0131 0.0118 0.0108 44.608
2 1.42 × 10− 3
6.46 × 10− 3 0.021603 0.0129 0.008 0.0095 26.990
3 1.46 × 10− 3
5.37 × 10− 3 0.015698 0.0128 0.0063 0.0083 20.877
4 1.53 × 10− 3
6.02 × 10− 3 0.028179 0.0129 0.0106 0.0102 33.808
5 1.34 × 10− 3
6.79 × 10− 3 0.018573 0.0130 0.006 0.0084 23.566
6 1.39 × 10− 3
5.51 × 10− 3 0.021782 0.0128 0.0108 0.0102 28.522
7 1.40 × 10− 3
6.26 × 10− 3 0.022549 0.0128 0.0123 0.0108 28.535
8 1.15 × 10− 3
3.40 × 10− 3 0.025138 0.0129 0.0107 0.0103 33.809
9 1.35 × 10− 3
5.54 × 10− 3 0.017423 0.0129 0.0051 0.0079 23.554

Appendix A4. Definitions of constants used in mathematical modelling of each metric

Parameter(s) Units Value Reference/remarks

P(i) W 400 Idle power

P(su) W 400 Setup power
P(l) W 425 MQL system power
P(Fp) W 100 Nanofluid preparation Power (Ultrasonic cleaner)
t(i) s 30 Idle time
t(su) s 150 Setup time
t(a) s 15 Air-cutting time
t(tc) s 60 Tool change time
t(l) s t(c) = t(l) lubrication time
t(d) (MQL) s 0 Fluid disposal time
x(e) $/kWh 0.105 Cost of electricity [55]
x(Wp) $/mm3 0.00138 Cost of the workpiece [56]
x(Fix) $/unit 14.4 Cost of the tool holder
x(CT) $ 0.48 Cost per cutting edge
x(F) $/L 7.21 Cost of the cutting fluid
x(Fp) $/sample 0.0002 Cost of the nanofluid preparation
x(Np) $/kg 721.5 Cost of nanoparticles (Graphene and Alumina)
x(Cp) $/mm3 5.91 × 10− 6 Cost of cleaning related process [57]
x(d) (MQL) $/L 0 Unit cost of disposal (fluid) [22]
x(d) (Flood) $/L 1.24 Unit cost of disposal (fluid) [22]
C(env) $/kg-CO2 0.057 Climate change bill [12]
CEF Kg-CO2/GJ 258.2 Carbon Emission Factor of Nanjing grid [13]
CF(Wp) kg-CO2/kg 9.8 Carbon footprints of workpiece material
CF(Fix) kg-CO2/kg 1.37 Carbon footprints of a fixture
CF(CT) kg-CO2 0.3787 Carbon footprints per cutting tool [58]
CF(F) kg-CO2/L 0.11 Carbon footprints of nanofluid preparation [59]
CF(Np) kg-CO2/kg 1520 Carbon footprints of nanoparticles [60]
Q(F) (MQL) L/h 0.30 Consumption rate of cutting fluid (biodegradable vegetable-based oil)
Q(F) (Flood) L/h 240 Consumption rate of cutting fluid (conventional emulsion)
Q(V) L – Used volume of cutting fluid in NFMQL and Flood assisted machining

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.rser.2021.110738.

16
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

References [29] Khan AM, Jamil M, Mia M, Pimenov DY, Gasiyarov VR, Gupta MK, et al. Multi-
objective optimization for grinding of AISI D2 steel with Al2O3 wheel under MQL.
Materials 2018. https://doi.org/10.3390/ma11112269.
[1] Lin B, Chen G. Energy efficiency and conservation in China’s manufacturing
[30] Khan Aqib Mashood, Jamil M, et al. Sustainable machining. Modeling and
industry. J Clean Prod 2018. https://doi.org/10.1016/j.jclepro.2017.10.286.
optimization of temperature and surface roughness in the milling of AISI D2 steel.
[2] National NBS. Bureau of statistics of China. 2019. http://data.stats.gov.cn/english
Ind Lubric Tribol 2018. https://doi.org/10.1108/ILT-11-2017-0322.
/easyquery.htm?cn=C01.
[31] Dietmair A, Verl A. A generic energy consumption model for decision making and
[3] Dahmus JB, Gutowski TG. An environmental analysis of machining. Manuf Eng
energy efficiency optimisation in manufacturing. Int J Sustain Eng 2009. https://
Mater Handl Eng 2004;2004:643–52. https://doi.org/10.1115/IMECE2004-62600.
doi.org/10.1080/19397030902947041.
[4] Santos JP, Oliveira M, Almeida FG, Pereira JP, Reis A. Improving the
[32] Balogun VA, Mativenga PT. Modelling of direct energy requirements in mechanical
environmental performance of machine-tools: influence of technology and
machining processes. J Clean Prod 2013;41:179–86. https://doi.org/10.1016/j.
throughput on the electrical energy consumption of a press-brake. J Clean Prod
jclepro.2012.10.015.
2011. https://doi.org/10.1016/j.jclepro.2010.10.009.
[33] Kennedy J, Eberhart R, Coello CAC, Pulido GT, Lechuga MS, Kennedy J, et al.
[5] Mori M, Fujishima M, Inamasu Y, Oda Y. A study on energy efficiency
Particle swarm optimization. Neural Networks. IEEE Int Conf; 1995. https://doi.
improvement for machine tools. CIRP Ann - Manuf Technol 2011;60:145–8.
org/10.1109/ICNN.1995.488968.
https://doi.org/10.1016/j.cirp.2011.03.099.
[34] Yang XS. Firefly algorithms for multimodal optimization. Lect Notes Comput Sci
[6] Davim JP. Design of optimisation of cutting parameters for turning metal matrix
(including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 2009. https://
composites based on the orthogonal arrays. J Mater Process Technol 2003. https://
doi.org/10.1007/978-3-642-04944-6_14.
doi.org/10.1016/S0924-0136(02)00946-9.
[35] AM Khan, M Jamil, K Salonitis, S Sarfraz, W Zhao NH. Multi-objective optimization
[7] Jamil M, Zhao W, He N. Sustainable milling of Ti–6Al–4V: A trade-off between
of energy consumption and surface quality in nanofluid SQCL assisted face milling.
energy efficiency, carbon emissions and machining characteristics under MQL and
Energies n.d. doi:doi.org/10.3390/en12040710.
cryogenic environment. J Clean Prod 2021. https://doi.org/10.1016/j.
[36] Ezugwu EO. Improvements in the machining of aero-engine alloys using self-
jclepro.2020.125374.
propelled rotary tooling technique. J Mater Process Technol 2007. https://doi.org/
[8] Iqbal A, Biermann D, Abbas H, Al-Ghamdi KA, Metzger M. Machining β-titanium
10.1016/j.jmatprotec.2006.03.112.
alloy under carbon dioxide snow and micro-lubrication: a study on tool deflection,
[37] Haddad Z, Abid C, Oztop HF, Mataoui A. A review on how the researchers prepare
energy consumption, and tool damage. Int J Adv Manuf Technol 2018. https://doi.
their nanofluids. Int J Therm Sci 2014. https://doi.org/10.1016/j.
org/10.1007/s00170-018-2267-4.
ijthermalsci.2013.08.010.
[9] Li L, Li C, Tang Y, Yi Q. Influence factors and operational strategies for energy
[38] He L, Shen J, Zhang Y. Ecological vulnerability assessment for ecological
efficiency improvement of CNC machining. J Clean Prod 2017;161:220–38.
conservation and environmental management. J Envir Manag 2018. https://doi.
https://doi.org/10.1016/j.jclepro.2017.05.084.
org/10.1016/j.jenvman.2017.11.059.
[10] GuanghuiZhou. Cutting parameter optimization for machining operations
[39] Lu H, Tian P, He L. Evaluating the global potential of aquifer thermal energy
considering carbon emissions. J Clean Prod 2018.
storage and determining the potential worldwide hotspots driven by socio-
[11] Yoon HS, Kim ES, Kim MS, Lee JY, Lee GB, Ahn SH. Towards greener machine tools
economic, geo-hydrologic and climatic conditions. Renew Sustain Energ Rev 2019.
- a review on energy saving strategies and technologies. Renew Sustain Energy Rev
https://doi.org/10.1016/j.rser.2019.06.013.
2015. https://doi.org/10.1016/j.rser.2015.03.100.
[40] Camposeco-Negrete C. Optimization of cutting parameters for minimizing energy
[12] Branker K, Adams D, Jeswiet J. Initial analysis of cost, energy and carbon dioxide
consumption in turning of AISI 6061 T6 using Taguchi methodology and ANOVA.
emissions in single point incremental forming - producing an aluminium hat. Int J
J Clean Prod 2013;53:195–203. https://doi.org/10.1016/j.jclepro.2013.03.049.
Sustain Eng 2012;5:188–98. https://doi.org/10.1080/19397038.2011.634033.
[41] Hu X, Chong HY, Wang X, et al. Sustainability perceptions of off-site manufacturing
[13] Li C, Tang Y, Cui L, Li P. A quantitative approach to analyze carbon emissions of
stakeholders in Australia. J Clean Prod 2019. https://doi.org/10.1016/j.
CNC-based machining systems. J Intell Manuf 2015. https://doi.org/10.1007/
jclepro.2019.03.258.
s10845-013-0812-4.
[42] Lv Q, Liu H, Yang D, Liu H. Effects of urbanization on freight transport carbon
[14] Yi Q, Tang Y, Li C, Li P. Optimization of CNC machine processing parameters for
emissions in China: Common characteristics and regional disparity. J Clean Prod
low carbon manufacturing. IEEE Int. Conf. Autom. Sci. Eng. 2013. https://doi.org/
2019. https://doi.org/10.1016/j.jclepro.2018.11.182.
10.1109/CoASE.2013.6654011.
[43] Velchev S, Kolev I, Ivanov K, Gechevski S. Empirical models for specific energy
[15] Tridech S, Cheng K. Low carbon manufacturing: characterization, THEORITICAL
consumption and optimization of cutting parameters for minimizing energy
models and implementation. Evolution 2008:1–14. https://bura.brunel.ac.uk/b
consumption during turning. J Clean Prod 2014;80:139–49. https://doi.org/
itstream/.
10.1016/j.jclepro.2014.05.099.
[16] Jeswiet J, Kara S. Carbon emissions and CESTM in manufacturing. CIRP Ann -
[44] Yi Q, Li C, Tang Y, Chen X. Multi-objective parameter optimization of CNC
Manuf Technol 2008. https://doi.org/10.1016/j.cirp.2008.03.117.
machining for low carbon manufacturing. J Clean Prod 2015;95:256–64. https://
[17] Tolouei-Rad M, Bidhendi IM. On the optimization of machining parameters for
doi.org/10.1016/j.jclepro.2015.02.076.
milling operations. Int J Mach Tool Manufact 1997. https://doi.org/10.1016/
[45] Li C, Li L, Tang Y, Zhu Y, Li L. A comprehensive approach to parameters
S0890-6955(96)00044-2.
optimization of energy-aware CNC milling. J Intell Manuf 2016:1–16. https://doi.
[18] Zhou L, Li J, Li F, Meng Q, Li J, Xu X. Energy consumption model and energy
org/10.1007/s10845-016-1233-y.
efficiency of machine tools: a comprehensive literature review. J Clean Prod 2016;
[46] Hegab H, Umer U, Deiab I, Kishawy H. Performance evaluation of Ti–6Al–4V
112:3721–34. https://doi.org/10.1016/j.jclepro.2015.05.093.
machining using nano-cutting fluids under minimum quantity lubrication. Int J
[19] Shin YC, Joo YS. Optimization of machining conditions with practical constraints.
Adv Manuf Technol 2018. https://doi.org/10.1007/s00170-017-1527-z.
Int J Prod Res 1992. https://doi.org/10.1080/00207549208948198.
[47] Kara S, Li W. Unit process energy consumption models for material removal
[20] Tipnis VA. Product life cycle economic models — towards a comprehensive
processes. CIRP Ann - Manuf Technol 2011;60:37–40. https://doi.org/10.1016/j.
framework for evaluation of environmental impact and competitive advantage.
cirp.2011.03.018.
CIRP Ann - Manuf Technol n.d.;40:463–466. doi:https://doi.
[48] Al-Ghamdi KA, Hussain G. On the CO2characterization in incremental forming of
org/10.1016/S0007-8506(07)62030-7.
roll bonded laminates. J Clean Prod 2017;156:214–25. https://doi.org/10.1016/j.
[21] Anderberg SE, Kara S, Beno T. Impact of energy efficiency on computer
jclepro.2017.04.047.
numerically controlled machining. Proc Inst Mech Eng Part B J Eng Manuf 2010;
[49] Jamil M, Khan AM, Hegab H, Gong L, Mia M, Gupta MK, et al. Effects of hybrid Al 2
224:531–41. https://doi.org/10.1243/09544054JEM1712.
O 3 -CNT nanofluids and cryogenic cooling on machining of Ti–6Al–4V. Int J Adv
[22] Priarone PC, Robiglio M, Settineri L. On the concurrent optimization of
Manuf Technol 2019. https://doi.org/10.1007/s00170-019-03485-9.
environmental and economic targets for machining. J Clean Prod 2018;190:
[50] Klocke F, Eisenblaetter G. Dry cutting. CIRP Ann - Manuf Technol 1997. https://
630–44. https://doi.org/10.1016/j.jclepro.2018.04.163.
doi.org/10.1016/S0007-8506(07)60877-4.
[23] Sidik NAC, Mohammed HA, Alawi OA, Samion S. A review on preparation methods
[51] Fister I, Yang XS, Brest J. A comprehensive review of firefly algorithms. Swarm
and challenges of nanofluids. Int Commun Heat Mass Tran 2014. https://doi.org/
Evol Comput 2013. https://doi.org/10.1016/j.swevo.2013.06.001.
10.1016/j.icheatmasstransfer.2014.03.002.
[52] Khan AM, Jamil M, Salonitis K, Sarfraz S, Zhao W, He N, et al. Multi-objective
[24] Devendiran DK, Amirtham VA. A review on preparation, characterization,
optimization of energy consumption and surface quality in nanofluid SQCl assisted
properties and applications of nanofluids. Renew Sustain Energy Rev 2016.
face milling. Energies 2019. https://doi.org/10.3390/en12040710.
https://doi.org/10.1016/j.rser.2016.01.055.
[53] Jeswiet J, Kara S. Carbon emissions and CESTM in manufacturing. CIRP Ann -
[25] Raja M, Vijayan R, Dineshkumar P, Venkatesan M. Review on nanofluids
Manuf Technol 2008;57:17–20. https://doi.org/10.1016/j.cirp.2008.03.117.
characterization, heat transfer characteristics and applications. Renew Sustain
[54] Skerlos SJ, Hayes KF, Clarens AF, Zhao F. Current advances in sustainable
Energy Rev 2016. https://doi.org/10.1016/j.rser.2016.05.079.
Metalworking Fluids research. Int J Sustain Manuf 2008. https://doi.org/10.1504/
[26] Rapeti P, Pasam VK, Rao Gurram KM, Revuru RS. Performance evaluation of
IJSM.2008.019233.
vegetable oil based nano cutting fluids in machining using grey relational analysis-
[55] Yi Q, Li C, Tang Y, Chen X. Multi-objective parameter optimization of CNC
A step towards sustainable manufacturing. J Clean Prod 2016;172:2862–75.
machining for low carbon manufacturing. J Clean Prod 2015. https://doi.org/
https://doi.org/10.1016/j.jclepro.2017.11.127.
10.1016/j.jclepro.2015.02.076.
[27] Laurent A, Olsen SI, Hauschild MZ. Carbon footprint as environmental
[56] Makeitform. Hayness 25 alloy. https://www.makeitfrom.com/material-pro
performance indicator for the manufacturing industry. CIRP Ann - Manuf Technol
perties/Annealed-R30605-Alloy. [Accessed 1 October 2018].
2010. https://doi.org/10.1016/j.cirp.2010.03.008.
[57] Priarone PC, Robiglio M, Settineri L, Tebaldo V. Modelling of specific energy
[28] Kishawy HA, Kannan S, Balazinski M. An energy based analytical force model for
requirements in machining as a function of tool and lubricoolant usage. CIRP Ann -
orthogonal cutting of metal matrix composites. CIRP Ann - Manuf Technol 2004.
Manuf Technol 2016;65:25–8. https://doi.org/10.1016/j.cirp.2016.04.108.
https://doi.org/10.1016/S0007-8506(07)60652-0.

17
A.M. Khan et al. Renewable and Sustainable Energy Reviews 139 (2021) 110738

[58] Rajemi MF, Mativenga PT, Aramcharoen A. Sustainable machining: selection of proposal of evaluation indicator for eco-efficiency). Trans Japan Soc Mech Eng Ser
optimum turning conditions based on minimum energy considerations. J Clean C 2006;72:923–8. https://doi.org/10.1299/kikaic.72.923.
Prod 2010;18:1059–65. https://doi.org/10.1016/j.jclepro.2010.01.025. [60] ICE. ICE. V2.0 - CarbonSolutions. Carbon Solut n.d. www.carbonsolutions.
[59] Narita H, Kawamura H, Chen L, Fujimoto H, Norihisa T, Hasebe T. Development of com/resources/ice v2.0 - jan 2011.xls. [Accessed 1 October 2018].
prediction system of environmental burden for machine tool operation (2nd report,

18