Sage Journals: Discover world-class research

Abstract

Manufacturing processes are among the most energy intensive on earth. As negative ecological and economic impacts increase, reducing energy consumption is becoming critically important. In this article, a comprehensive overview of energy-saving strategies and opportunities for increasing energy efficiency in manufacturing operations is presented, with a focus on metal cutting processes. The issues and approaches involved in energy efficiency of machine tools and machining operations are reported in the literature and a structured research methodology is proposed for this purpose including prediction and modelling of machine energy consumption, determining the relationship between process energy consumption and process variables for material removal processes and optimization of cutting parameters in order to reduce energy consumption. Numerous techniques for increasing energy efficiency in manufacturing processes are identified and summarized, strengths and weaknesses of previous studies are discussed and potential avenues for future research are suggested.

Keywords

Energy efficiency machining processes parameter optimization machine tool

Introduction

The manufacturing sector comprises some of the largest energy consumers and carbon emitters in the world, accounting for about 33% of primary energy use and 38% of CO₂ emissions globally^1,2 and all associated ecological impacts affiliated with it.³ On the other hand, the rising cost of energy also is affecting the manufacturing industry economically. Reducing the amount of consumed energy and implementing more energy-efficient manufacturing processes can significantly enhance performance of processes and reduce its undesirable impacts. Thus, scholars strongly agree that performing an inclusive review of manufacturing processes to determine overall energy demand is essential.⁴ Machining is one of the most fundamental and important manufacturing processes that is widely used in manufacturing industries. As shown in Figure 1, machining is also one of the main energy consuming practices in the manufacturing sector and has been identified as a main target for energy reduction in recent years. Most existing studies on energy reduction in the manufacturing sector are focused on developing the most energy-efficient machining processes.^5,6

Figure 1.

Machine drive electricity use as a percentage of delivered energy use by each industry.¹

In order to create techniques to save energy during machining processes and increase energy efficiency, achieving a credible prognostication of energy consumption of machine tools by analysing the power consumption of their various components is essentially required. Using optimized cutting parameters is an effective technique for controlling energy consumption during manufacturing processes. In this article, the results of studies in which optimal cutting parameters have been determined through the use of various optimization methods are synthesized.

Motivation for review

Increasing demand for energy in manufacturing and machining, which are major energy consuming processes, is a significant factor underlying current environmental and economic challenges. While a major goal of scholars in the field of energy-efficient manufacturing is achieving minimal energy consumption,⁷ they also tend to seek solutions that are advantageous from both an economic and environmental perspective.⁸ As a result, recent studies performed by many researchers have focused on energy-related efforts. In this article review, the existing methodologies related to energy efficiency and reduction in energy consumption provide an arranged classification of these works under a specified framework which will be useful for identification of the barriers of the developed techniques and also future challenges concerning energy efficiency of machining processes.

Research methodology

To perform this literature review, journal articles and conference proceedings related to the energy efficiency of manufacturing processes and the optimization of cutting parameters were compiled, and a total of 101 articles matching the scope of our survey were identified. A structured research methodology was applied in order to categorize studies related to the energy efficiency of manufacturing processes into three main groups: theoretical modelling methods, empirical modelling methods and the optimization of cutting parameters. The research methodology is depicted in Figure 2. This process enabled numerous techniques for increasing energy efficiency in manufacturing processes to be synthesized and strengths, weaknesses, challenges and limitations of previous studies to be identified.

Figure 2.

Research methodology for this review article.

Categorical literature review

Energy consumption models

Analysing the energy consumption of machine tools is a critical first step towards reducing the total amount of energy consumed during machining operations. An energy monitoring system is indispensable for obtaining information about energy flow. Kara et al.⁹ synthesized the latest improvements in electricity metering and monitoring systems and related standards.

One of the most important developments in this area is a software-based approach for automatic energy ratiocination. Vijayaraghavan and Dornfeld¹⁰ used a complex event processing methodology to monitor and analyse the energy consumption of machine tools and other manufacturing equipment. They developed a framework for temporally analysing the energy consumption of machine tools, which can be extended to other types of environmentally relevant data streams in manufacturing systems in order to improve their environmental efficiency.

Le et al.¹¹ created an algorithm to process real-time energy data in order to distinguish various operational states, thereby decreasing the number of sensors required. Oliver et al. categorized machining processes as either steady state or transient state and developed an energy monitoring approach for machine components. This approach has been used to monitor the energy consumption of spindle and feed axes.¹²

Among the developed energy monitoring techniques, the most promising are automated monitoring systems. Automated monitoring techniques can be used with complex systems where manual monitoring techniques cannot be used.

In various power consumption modelling studies, researchers tried to decompose and model cutting power. Power consumed by a tool can be measured by measuring the tool’s power consumption using a watt meter installed on the machine or by measuring the cutting forces acting on the tool. A number of experiments have focused on predicting and modelling the cutting force¹³ using analytic¹⁴ or numerical¹⁵ analysis.

Kishawy et al.¹⁶ introduced an energy-based analytical model that predicts the orthogonal cutting force for machining composite materials. Palanisamy et al.¹⁷ proposed a dynamic cutting force model including tangential and normal forces of an end-milling process.

Huang and Liang¹⁸ modelled cutting forces under different cutting conditions. Linear regression¹⁹ and fuzzy logic²⁰ were also used to model cutting force. Researchers have also attempted to model cutting power using cutting force models. Table 1 provides a summary of works in which power consumed during cutting has been decomposed and modelled.

Table 1.

A summary of cutting power prediction models.

Author	Model
Astakhov and Xiao²¹	$P_{machining} = P_{Pd} + P_{fR} + P_{fF} + P_{ch} + P_{mn - ce}$ where $P_{Pd}$ is the power consumed for plastic deformation, $P_{fR}$ is the power consumed between the tool and chip, $P_{fF}$ is the power consumed between the tool and work material, $P_{ch}$ is the power for formation of new surfaces and $P_{mn - ce}$ is the power used by minor cutting edges
Xu et al.²²	$P_{machining} = K_{TC} MRR + K_{TE} \overset{\cdot}{A}$ where $K_{TC}$ is the coefficient of cutting energy, $K_{TE}$ is the coefficient of the tool edge force and $A$ is the change in the rate of interface between the tool and work material
Cuppini et al.²³	$P_{machining} = α \bar{VB} + β$ where $α$ and $β$ are the empirical constants, and VB is the flank wear of the tool
Shao et al.²⁴	$P_{machining} = ZnD a_{p} [K {\bar{h}}^{- c} f_{z} (\cos (φ_{in}) - \cos (φ_{in} + \emptyset)) + μ H \bar{VB \emptyset}] / 2$ where Z is the number of teeth, D is the diameter of the cutting tool, K is the cutting force constant, h is the mean chip thickness, c is the chip thickness constant, $μ$ is the friction coefficient between the work material and tool, H is the Brinell hardness, $\emptyset$ is the immersion angle, $φ_{in}$ is the cutting angle when a cutting tooth enters into the work material and VB is tool-flank wear

Astakhov and Xiao²¹ modelled cutting power using cutting force elements: the power consumed for plastic deformation of the layer being removed, the power consumed at the tool–chip interface, the power consumed at the tool–workpiece interface and the power consumed in the formation of new surfaces. This model is quite practical; total cutting power can be calculated simply and the contribution of each element is delineated.

Xu et al.²² modelled cutting power based on motor spindle power using cutting energy coefficients, and the accuracy of the model has been shown. Cuppini et al.²³ modelled cutting power based on tool wear, assuming a linear relationship between cutting power and tool wear.

Furthermore, Shao et al.²⁴ provided a model for calculating the mean cutting power in milling operations. Model verification experiments have been performed, and the mean cutting power counterbalances the inherent fluctuations in the measured cutting power signals.

Furthermore, several scholars have decomposed energy elements. Saidur²⁵ reviewed the energy consumption of machine tool motors, and Abele et al.²⁶ focused on the machine tool spindle as the representative unit and also looked for the possible improvements in energy efficiency.

In recent years, efforts to model the energy consumption of machine tools have increased considerably. Related studies and methodologies can be grouped into two broad categories: theoretical models and empirical models.

Theoretical modelling

A large number of studies involve the development of mathematical models to predict the electrical energy requirements of machine tools during machining operations. De Filippi et al.²⁷ were the first researchers to investigate energy efficiency. They gathered and analysed data associated with different operations from numerically controlled (NC) machine tools. They found that the energy consumed by machine tools during machining is considerably greater than the energy consumed during chip formation.

A precursor study that laid the foundation for determining the energy consumption rate was carried out by Kordonowy,²⁸ who collected power consumption data for six different machines and classified their subunits. In addition, Kordonowy classified power consumption values according to their operational characteristics into two groups of variable and constant. Power drawn by subunits such as computers, fans, servo motors, hydraulic pumps and unloaded motors, which are activated as soon as the machine starts up, is classified as constant part where the variable part is the power associated with machining processes and material removal rate. Gutowski et al.²⁹ showed that the energy consumed by the material removal process is much lower than the energy consumed by necessary auxiliary units in order to maintain the metal cutting process, using the monitoring system developed by Dahmus and Gutowski.³⁰ Figure 3 shows the energy used as a function of production rate for an automobile production machining line. It is noticeable that the energy consumed while the machine is idling far exceeds the energy consumed during machining processes.

Figure 3.

Energy used as a function of production rate for an automobile production machining line.³¹

Devoldere et al.³² also analysed the energy consumption of machine tools and contributions of subunits on a five-axis milling machine and showed that the machine consumes about 1.7 kW of energy in idle or standby mode. Gutowski et al.²⁹ pioneered the modelling of energy consumption to predict the energy consumed by machining processes. They assumed that a considerable amount of energy is required to start up a machine and maintain ready status. Additionally, energy demand fluctuates based on the rate of material processing. Therefore, they proposed a model in which the energy expended during the machining process could be calculated based on specific cutting energy. Although this model has paved the way for many other studies, it does not account for energy consumed by auxiliary subunits, which is not fixed and changes during machining operations. No experimental studies have been performed to measure the consistency of this model. Rajemi et al.³³ also modelled the energy consumed to produce a part during turning operations based on specific cutting energy. This model is focused on determining and optimizing the energy footprint not direct energy demand. An equation was developed to determine the optimum tool life for minimum energy in this study and also optimum cutting velocities were evaluated for this purpose. They also reported that machining at lower production rates decreases the rate of actual energy required for machining processes. Mativenga and Rajemi³⁴ improved Rajemi et al.’s³³ methodology by taking spindle power consumption into consideration.

In complex NC machine tools, peripheral subunits account for more than 30% of total energy consumption, and therefore, peripheral subunits could play a significant role in reducing the energy consumption of machine tools.³⁵ Since the total energy consumption of a machine is greater than the energy consumed during the cutting process, the energy consumed by a peripheral subunit must be taken into consideration. A number of studies have been performed for this purpose.

Figure 4 demonstrates the power profile of a turning process presented by Li et al.³⁶ In this study, Li et al. divided the total power consumption of a machine tool into four main groups:

Fixed power: power demand of all activated machine components ensuring the operational readiness of the machine.

Operational power: power demand to distinctively operate components enabling cutting as performed in air cuts.

Tool tip power: power demand at the tool tip to remove the workpiece material.

Unproductive power: power converted into heat mainly due to friction during material removal.

Figure 4.

Power profile of a turning process.³⁶

Li et al. also classified the fixed energy consumption of a machine tool, which is shown in Figure 5.

Figure 5.

Average fixed energy breakdown of a reviewed machine tool.³⁶

He et al.³⁷ modelled the energy consumed in machining operations by associating it with NC commands. They developed the model by identifying energy consuming subunits and determining their energy consumption amounts by observing the subunits’ commissioning times. This model is not a general model for all machine tools and it also fails to model the energy consumption related to the number of tool changes and spindle speed.

Salonitis and Ball³⁸ categorized the total energy consumption of machine tools into process energy, which is the energy consumed during the cutting process, and peripheral energy, which is the energy consumed by peripheral subunits. They also divided peripheral energy into background energy, which is the energy consumed by a machine tool in ready position, irrespective of whether a cutting process is being performed or not, and load energy, which is the energy drawn during the cutting process and depends on cutting parameters.

Calvanese et al.³⁹ also modelled energy consumption based on functional modules including the spindle, axis, chillers, tool change system, auxiliary components and cutting process, but did not consider the effect of tool wear, multiple axis movement and tool tip energy.

Aramcharoen and Mativenga⁴⁰ suggested a model for calculating the total energy consumption of each cutting process, considering the energy consumed in ready position, and by tool changes, spindle runs, material removal, coolant, table movement and cutting feed rate. To measure the consistency of the model, they compared the predicted energy consumed with the energy measured in the experiment and found a 5% difference. Aramcharoen and Mativenga also analysed the effect of alternate tool paths on energy consumption.

Uluer et al.⁴¹ proposed a model for calculating the energy consumption of a turning part by analysing energy consumption at the system level and dividing it into direct and indirect energy. Unlike other models, metal cutting energy is calculated by multiplying the volume of the chip removed by the specific cutting energy, thereby providing a consistent method for component-based modelling. They did not perform any experiments to validate the suggested model.

To model the overall power consumption of a machine tool, Mori et al.⁴² divided a machining process into a number of various operations: positioning the spindle after tool change and acceleration, metal cutting, returning the spindle to the tool exchange location once metal cutting is complete and stopping the spindle. In the study, Mori et al. measured the energy consumption of different machining operations, including face milling, shoulder milling and drilling under various cutting conditions. They determined the optimal cutting conditions to reduce energy consumption by performing Taguchi techniques. In this model, the change in energy consumed by the tool tip based on the material being processed, the energy expenditure of the feed axis and the change in energy consumed by the spindle based on the spindle speed are not taken into account. Applying the suggested methods did result in reduced energy consumption, but a verification study between the theoretical model and experimental data has not yet been performed.

Diaz et al.⁴³ modelled the total energy consumption of a milling machine tool during a milling operation by measuring the energy drawn by the machine when cutting the air. In this study, a number of experiments with varying depth of cut and width of cut were conducted to demonstrate the effect of material removal rate on the energy consumption of a three-axis machining centre which is shown in Figure 6. They investigated the relationship between power consumption and machining time and confirmed that although power consumption increases with higher material removal rates, the total energy consumption of the machining centre does not increase due to reduction in total machining time. They also examined the effect of workpiece material on the amount of power drawn by the machine. This model’s handicap is its failure to consider neither the power drawn by the machine when it is in ready position nor the power consumption of auxiliary units. Furthermore, the theoretical model has not been compared against experimental data in order to prove the model’s consistency. In a previous study, Diaz et al.⁴⁴ analysed the effect of cutting speed on energy consumption by comparing high-speed and conventional machining and found that reduced machining time in high-speed machining makes it more energy efficient.

Figure 6.

Energy and power demand as a function of MRR.⁴³

Kong et al.⁴⁵ developed a model for estimating the energy consumption of a computer numerical control (CNC) machine tool by categorizing energy consumption as constant, run-time or cutting. They also investigated the effects of different tool paths on the energy consumption of a machine tool using a process analysis software tool with a web-based application programming interface (API). However, they did not consider the effects of tool changes, individual motion of the feed axis and coolant in this model.

It is available to gather data about the specific machining operations in modern CNC.⁴⁶ Moreover, many computer-aided design (CAD)/computer-aided manufacturing (CAM) applications include energy assessment tools⁴⁷ that provide overall energy estimates. Avram and Xirouchakis¹² developed a model that calculates energy consumed in a metal cutting process based on feed rates and spindle speed values from output files of the CATIA v.5 rev.15 program related to various tool paths for 2.5-axis milling operations. Calculations are performed by considering not only feed rates and spindle speed values but also information about the number of cutting mouth, helix angle, side angle, cutting depth, cutting width and specific cutting energy for the selected material. Energy consuming units and the amount of consumed energy are determined by performing milling operations. The consumed energy is divided into four main groups: constant, feed axis, spindle and auxiliary energy. Thus, this model shifts away from classifying energy consumption as either auxiliary or metal cutting. In this model, metal cutting energy consumption is covered by the spindle and axis energy consumption (which also includes energy drawn by these units in ready position). Neugebauer et al.⁴⁸ also modelled the energy consumption of machine tools by identifying energy consumers in their system-level study in order to use machine tools more efficiently.

Hu et al.⁴⁹ modelled the energy consumption of machine tools through the energy consumption of the spindle and divided the power drawn by the spindle into three sub-groups: idle power, cutting power and additional load loss. Hu et al.⁵⁰ also modelled the power consumption resulting from additional losses as a quadratic function. Hu et al.⁴⁹ improved the presented model using this function and the experiments have been performed to validate it. Data obtained from a torque sensor are multiplied by the angular velocity of the spindle to calculate the total power drawn by the spindle, which is compared with data obtained from power sensor. In direct comparisons based on six tests conducted under different cutting conditions, the error rates ranged from 16.75% to 24.09%. Following the proposed procedure, additional loss coefficients of the turning machine were determined. Removing the additional losses from the power drawn during cutting resulted in an error rate of 3%. This energy model provides information such as real-time energy efficiency, instantaneous power drawn, total energy consumed during the operation, energy used during metal cutting and operation time based on computer software outputs. This model is used to classify rather than predict energy consumption, because additional losses are extrapolated from the amount of online power drawn.

Balogun and Mativenga⁵¹ proposed their own energy consumption model within the scope of Cooperative Effort in Process Emission (CO2PE) using Kellens et al.’s⁵² proposed methodology. In this model, the energy consumed is divided into three groups: basic, ready for cutting and metal cutting energy. Basic energy is defined as the energy required to start up and activate computer units, lighting, cooling fans, lubrication and energy consumed by unloaded motors and so on. The influences of feed rate and undeformed chip thickness were also analysed in this study. Table 2 provides a summary of the developed models.

Table 2.

A summary of developed theoretical models.

Reference	Mathematical model	Remark
Gutowski et al.²⁹	$E = (P_{0} + kQ) t$	A precursor work to determine energy consumption
	$E = (P_{0} + kQ) t$	Does not provide information about energy consumption of auxiliary subunits
	$P_{0}$ : power consumed by machine, k: specific cutting energy, Q: material removal rate, t: cutting time	No model validation studies
Rajemi et al.³³	$E = P_{0} t_{1} + (P_{0} + kQ) t_{2} + P_{0} t_{3}) (t_{2} / T) + y_{E} (t_{2} / T)$	A new model and methodology for optimizing the energy footprint
Rajemi et al.³³	$P_{0}$ : power consumed by machine, $t_{1}$ : machining setup time, $t_{2}$ : actual cutting time, $t_{3}$ : tool change time, $T$ : tool life, $y_{E}$ : energy footprint for tool edge, k: specific cutting energy, Q: material removal rate	The most effective parameters for minimizing energy consumption are identified
Mori et al.⁴²	$E = P_{1} (T_{1} + T_{2}) + P_{2} (T_{2}) + P_{3} (T_{3})$	Does not provide information about the power calculation of each process.
	$P_{1}$ : constant power during machining, $P_{2}$ : power for cutting by spindle and servo motor, $P_{3}$ : power to position the work to accelerate/decelerate the spindle to the specified speed, $T_{1}$ : cycle time during non-cutting state, $T_{2}$ : cycle time during cutting state, $T_{3}$ : time required to position the work and accelerate the spindle	Effects of feed axis, spindle speed and tool life are not considered in the model
		No model validation studies
Diaz et al.⁴³	$E = (P_{cut} + P_{air}) \cdot Δ T$	Introduced an interesting approach by considering air cut power consumption
	$E = (P_{cut} + P_{air}) \cdot Δ T$	Does not provide information about energy consumption of auxiliary subunits
	$P_{cut}$ : power demand for cutting, $P_{air}$ : power demand for air cut, $Δ T$ : processing time	Requires measuring the air-cutting time for the toolpath
		No model validation studies
Kong et al.⁴⁵	$E = E_{const} + E_{run - rime - transient} + E_{run - time - steady} + E_{cut}$	Investigation of the effect of different tool paths on energy consumption
Kong et al.⁴⁵	$E_{const}$ : energy consumed not related to machining, $E_{run - rime - transient}$ : energy consumed by spindle, machine axis and tool changes when accelerating or decelerating to reach specified values, $E_{run - time - steady}$ : energy consumed by spindle, machine axis and tool changes when the spindle motor and axis drives maintain a specified value, $E_{cut}$ : energy consumed by material removal action	Does not provide information about energy consumption of coolant and tool changes
He et al.³⁷	$E = E_{spindle} + E_{feed} + E_{tool} + E_{coolant} + E_{fix}$	Not a general model for all machine tools
He et al.³⁷	$E_{spindle}$ : energy consumption of spindle, $E_{feed}$ :energy consumption of axis feed, $E_{tool}$ : energy consumption of tool change, $E_{coolant}$ : energy consumption of coolant, $E_{fix}$ : energy consumption of fan motor and servo system	Fails to model the energy consumption related to the number of tool changes and spindle speed
Calvanese et al.³⁹	$E = E_{fixed} + E_{axes} + E_{axis chillers} + E_{spindle} + E_{spindle chiller} + E_{chip conveyor} + E_{tool change} + E_{pallet clamp}$	The effects of tool wear, multiple axis movement, and tool tip energy are not taken into consideration
Calvanese et al.³⁹	$E_{fixed}$ : the energy required when machine is switched on, $E_{axes}$ : the energy required by the axis, $E_{axis chillers}$ : the energy consumed by axis chillers, $E_{spindle}$ : the energy of both the cutting power and the spindle losses, $E_{spindle chiller}$ : the energy consumed by spindle chillers, $E_{chip conveyor}$ : the energy consumed when chip is removed, $E_{tool change}$ : the energy consumed during tool change, $E_{pallet clamp}$ : the energy consumed during clamping
Balogun and Mativenga⁵¹	$E = E_{b} + E_{r} + P_{tc} t_{tc} [INT (t_{2} / T) + 1] + P_{air} t_{air} + (P_{s} + P_{cool} + k \overset{\cdot}{v}) t_{c}$	Reported the effect of machine tool selection and geographic location on energy and carbon footprint
Balogun and Mativenga⁵¹	$E_{t}$ : the direct total energy requirement, $E_{r}$ : ready state power, $P_{tc}$ : power demand for tool change $t_{tc}$ : time required for tool change, $T$ : tool life, $P_{air}$ : the average power requirements for a non-cutting approach and retraction moves over the component, $t_{air}$ : the total duration of non-cutting moves, $P_{s}$ : $P_{cool}$ : coolant pumping power requirements, $k$ : specific cutting energy, $\overset{\cdot}{v}$ : the rate of material processing, $t_{c}$ : cutting time	Model validation studies have been performed
Salonitis and Ball³⁸	$E_{total} = E_{process} + E_{background} + E_{load}$	Does not provide information about energy consumption of auxiliary subunits
Salonitis and Ball³⁸	$E_{process}$ : energy required for the physical process to occur, $E_{background}$ : energy consumed by the machine tool in ready position, irrespective whether a cutting process is being performed, $E_{load}$ : energy drawn during cutting process based on cutting parameters
Uluer et al.⁴¹	$E_{part} = E_{th} + E_{aux - var} + E_{aux - const} + E_{handling} + E_{indirect}$	No model validation studies
Uluer et al.⁴¹	$E_{th}$ : theoretical energy consumed during the actual metal removal processes, $E_{aux - var}$ : variable auxiliary energy, which may change during the operation of the machine tool, $E_{aux - const}$ : constant auxiliary energy consumed by auxiliary components of the machine tools even when the machine is in standby mode, $E_{handling}$ : energy consumed by automated material handling equipment, $E_{indirect}$ : allocated indirect energy consumed by services to maintain the production environment	No model validation studies
Avram and Xirouchakis¹²	$E_{DE} = E_{aY} + E_{SY} + E_{dY} + E_{run} + E_{cut}$	Fails to model the energy consumption of auxiliary components
Avram and Xirouchakis¹²	$E_{DE}$ : total energy consumption, $E_{aY}$ : energy consumed for positive acceleration of spindle in y-axis, $E_{SY}$ : steady-state energy consumption of spindle in y-axis, $E_{dY}$ : energy consumed for negative acceleration of spindle in y-axis, $E_{run}$ : steady-state energy consumption of spindle, $E_{cut}$ : energy consumption of spindle during metal cutting process	Verification studies have been performed and accuracy of about 75%–85% has been obtained
Neugebauer et al.⁴⁸	$E_{total} = E_{use, prime} + E_{use, \sec} + E_{loss}$	Fails to model the energy consumption of auxiliary components
Neugebauer et al.⁴⁸	$E_{use, prime}$ : energy consumed by auxiliary subunits, $E_{use, \sec}$ : energy consumed for metal cutting processes buy machine tool components	The effect of machining operations such as dry machining, roughing or finishing is advertised
Hu et al.⁵⁰	$P = P_{u} + (1 + a_{0}) P_{c} + a_{1} P_{c}^{2}$	Fails to model the energy consumption of auxiliary components
Hu et al.⁵⁰	$P = P_{u} + (1 + a_{0}) P_{c} + a_{1} P_{c}^{2}$	Requires more detailed information about power consumption of auxiliary components
	$P_{u}$ : idle power, $P_{c}$ : cutting power, $a_{0} and a_{1}$ : additional load loss coefficients	Verification studies have been performed

Empirical modelling

Empirical models have provided reliable predictions of the energy use of machine tools and characterizations of the relationship between energy consumption and cutting parameters. Draganescu et al.⁵³ introduced a detailed model of specific energy consumption based on cutting parameters during milling operations. They also graphed the relationships between efficiency and cutting parameters, tangential cutting forces and cutting parameters and specific energy consumption and cutting parameters using the same cutting tool, material and machine while varying the cutting parameters. Specific consumed energy as a function of milling parameters is shown in Figure 7. Although this model is a good basis for further study, it only calculates energy consumed for the cutting process and does not take into account the energy consumption of subunits. Furthermore, this is not a generic model of other processes and machine tools.

Figure 7.

Specific consumed energy as a function of milling parameters.⁵³

Kara and Li⁵⁴ developed a unit process energy consumption model using experimental methods to observe energy consumption related to process parameters in order to determine the relationship between these parameters and energy consumption. They identified constants specific to each machine tool and revealed an empirical model based on these machine-specific coefficients to reliably predict unit process energy consumption. The experiments were designed based on process variables and response surface methodology (RSM). A monitoring platform was constructed using the LabVIEW™ programming interface and statistical analysis of unit process energy consumption was performed using SPSS software. Model validation tests have been performed on four different milling and turning machine tools. A comparison of model estimates and measured energy amounts revealed 90% consistency. However, the estimated energy does not include start-up, standby, clamping and positioning energy consumption. In the study, Specific Energy Consumption (SEC) formulation was derived separately for the operations with and without coolant, but it has not been demonstrated how changes taking place in other auxiliary subunits affect total energy consumption. The results of this study show that less energy is consumed when material removal rates are higher. Guo et al.⁵⁵ developed an empirical model with respect to cutting parameters and determined optimal cutting parameters in order to obtain minimum energy consumption and precise surface finishes.

Yoon et al.⁵⁶ further categorized the energy consumption of machine tools into basic energy, spindle energy, stage energy and material removal energy. Basic energy represents the energy consumption of auxiliary subunits and control systems, which are assumed to be constant. Material removal energy is the energy consumed during the material removal process. Stage energy and spindle energy, which are the energy required for spindle rotation and stage movement, have a power relationship based on rotational speed and feed rate. They developed an empirical model by performing experimental studies and fitting the observed data. Yoon et al.⁵⁷ also classified the energy consumed during different stages of milling operations, and material removal power was shown to depend on cutting parameters and tool wear, as shown in Figure 8.

Figure 8.

Material removal power with respect to the tool wear under the specified conditions.⁵⁷

Pervaiz et al.⁵⁸ used finite element modelling simulation to model energy consumption. Experimental verification revealed that the experimental results approximated measured power consumption values quite well. Dietmair and Verl⁵⁹ presented an energy consumption model which was improved later.⁶⁰ This model predicts energy consumption based on different design and operations strategies. Larek et al.⁶¹ also presented a power consumption prediction approach in which all components of a machine tool are assumed to be either ‘on’ or ‘off’. Table 3 provides a summary of empirical models of machine tool energy consumption.

Table 3.

A summary of developed empirical methods.

Reference	Model	Process	Work material
Draganescu et al.⁵³	$E_{cs} = (π D F_{t}) / (3.672 \times 10^{6} s_{z} tBz μ), E_{c} = Y E_{cs}$	Milling	ATSi10Mg
Draganescu et al.⁵³	D: diameter of the mill, $s_{z}$ : feed per tooth, t: depth of milling, B: contact length of the milling tool, z: number of teeth of the milling tool, $F_{t}$ : tangential component of the cutting force, $E_{cs}$ : specific consumed energy, Y: total volume of material removed, $E_{c}$ : consumed energy	Milling	ATSi10Mg
Kara and Li⁵⁴	$E = SEC \cdot cutting volume$	Milling	Brass
	$SEC = C_{0} + (C_{1} / MRR)$	Turning	Mild steel
	SEC is the specific energy consumption, $C_{0}$ and $C_{1}$ are the machine-specific coefficients and MRR is the material removal rate	Turning	High tensile
Guo et al.⁵⁵	$SEC = C_{0} \cdot v_{c}^{α} \cdot f_{r}^{β} \cdot a_{p}^{γ} \cdot d^{φ} + ((C_{1}) / (v_{c} \cdot f_{r} \cdot a_{p}))$	Turning	Steel, aluminium
Guo et al.⁵⁵	$v_{c}$ : cutting speed, $f_{r}$ : feed rate, $a_{p}$ : depth of cut, d: final workpiece diameter (mm) in the turning operation, $α$ , $β$ , $γ$ , $φ$ , $C_{0}$ and $C_{1}$ are the empirical constants	Turning	Steel, aluminium
Yoon et al.⁵⁶	$E_{t} = E_{basic} + E_{stage} + E_{spindle} + E_{machining}$	Drilling	PCB
Yoon et al.⁵⁶	$E_{basic}$ : energy consumed when turning on the machine, $E_{stage}$ : energy consumed when moving the stage, during the air-cutting state, $E_{spindle}$ : energy consumed by rotating the spindle, $E_{machining}$ : additional process energy consumed in material removal	Drilling	PCB

PCB: printed circuit board.

The most important articles pertaining to the modelling of energy consumption of machine tools using either theoretical or empirical methods are also provided in Table 4, ranked based on the number of citations. A chronological distribution of these articles is provided in Figure 9.

Table 4.

Most cited articles related to the modelling of machine tool energy consumption.

Rank	Reference	Journal or proceedings	Method	No. of citations
1	Gutowski et al.²⁹	13th CIRP International Conference on Life Cycle Engineering	Theoretical modelling	327
2	Dahmus and Gutowski³⁰	2004 ASME International Mechanical Engineering Congress (IMECE) and RD&D Expo	Theoretical modelling	314
3	Rajemi et al.³³	Journal of Cleaner Production	Theoretical modelling	137
4	Kara and Li⁵⁴	CIRP Annals – Manufacturing Technology	Empirical modelling	129
5	Mori et al.⁴²	CIRP Annals – Manufacturing Technology	Theoretical modelling	120
6	Diaz et al.⁴³	18th CIRP International Conference on Life Cycle Engineering	Theoretical modelling	119
7	Draganescu et al.⁵³	Journal of Materials Processing Technology	Empirical modelling	116
8	Dietmair and Verl⁶⁰	International Journal of Sustainable Engineering	Empirical modelling	116
9	Avram and Xirouchakis¹²	Journal of Cleaner Production	Theoretical modelling	98
10	Hu et al.⁴⁹	Journal of Cleaner Production	Theoretical modelling	68

Figure 9.

Chronological distribution of articles related to the modelling of energy consumption.

Optimization of cutting parameters

Design of experiments

It has long been known that optimizing the economics and quality of machining processes which are determined by productivity, total energy consumption, total cost and other criteria requires the selection of cutting parameters such as feed rate, cutting speed and depth of cut in machining processes.⁶² Understanding the energy consumption of machine tools as a function of cutting processes is necessary in order to improve energy efficiency.⁴³ The statistical design of experiments (DOEs) is the process of planning experiments so that the appropriate data can be analysed by statistical methods and yield valid and objective conclusions.⁶²

A number of studies have been performed to optimize machining processes by considering cutting parameters and employing DOE methods such as factorial design, RSM and Taguchi methods. Thomas et al.⁶³ used a full-factorial design involving six factors where cutting speed, feed rate and depth of cut were the primary factors used to analyse the effects of cutting conditions, tool parameters and cutting force. Yang and Tarng⁶⁴ were the first to obtain optimal cutting parameters using the Taguchi method. In this study, the optimal cutting parameters were found by creating an orthogonal array, calculating the signal-to-noise (S/N) ratio, and performing an analysis of variance (ANOVA) for the process of turning S45C steel bars using tungsten carbide cutting tools. The authors also identified the cutting parameters that influence cutting performance in turning operations. Experimental studies have been performed in order to verify the applicability of this method. Several studies have been carried out to minimize the power consumption by obtaining optimized cutting parameters using Taguchi techniques. Fratila and Caizar⁶⁵ also used the Taguchi methodology in their work to optimize the cutting parameters in face milling in order to minimize power consumption. An orthogonal array, S/N ratio and ANOVA were employed to investigate the effect of cutting parameters when machining AlMg₃ (EN AW 5754) with a high-speed steel (HSS) tool. The experiments were performed under different conditions: dry cutting, minimal quantity lubrication and flood lubrication. The authors found that using the Taguchi method requires a minimum number of trials compared to a full-factorial design. Adinarayana et al.⁶⁶ also performed power optimization using Taguchi techniques in turning 4340 alloy steel using a chemical vapour deposition (CVD) cutting tool. In this study, Taguchi’s L27 orthogonal array was used to design the experiments and ANOVA was employed to analyse the effects of cutting parameters. Babu et al.⁶⁷ performed similar optimization experiments with an extruded aluminium shaft on a CNC lathe. It was noticed that the depth of cut and feed rate primarily affect power consumption. Kulkarni et al. also studied the parameter optimization for a CNC turning operation of AISI 1040 steel using the Taguchi methodology in order to minimize power consumption. In this study, Taguchi’s L27 orthogonal array and ANOVA were used to investigate the influence of nose radius and cutting fluid in addition to cutting speed, feed rate and depth of cut; the optimal values were obtained.⁶⁸

Bhattacharya et al.⁶⁹ and Hanafi et al.⁷⁰ tried to achieve the best surface roughness with minimum cutting power. Bhattacharya et al. performed an experimental study to analyse the contribution and effects of cutting speed, feed rate and depth of cut on surface roughness and power consumption during high-speed machining of AISI 1045 steel based on Taguchi techniques. An orthogonal array and ANOVA were employed for this purpose. They concluded that cutting speed most affects power consumption; they consequently optimized cutting parameters to minimize power consumption.⁶⁹

Hanafi et al. employed Taguchi’s orthogonal array, S/N ratio and grey relational analysis to achieve best surface roughness with minimum cutting power during dry turning of PEEK-CF30 using TiN tools. They concluded that the depth of cut most affects power consumption, followed by cutting speed and feed rate.³⁴

The Taguchi methodology was also used by Camposeco-Negrete when rough turning AISI 6061 T6 in order to minimize energy consumption. To analyse the influence and contributions of depth of cut, feed rate and cutting speed on energy consumption, the author employed an orthogonal array, S/N ratio and ANOVA. Cutting power, cutting energy, power consumed and energy consumed during the machining process were optimized and different values of cutting parameters were obtained for each.⁷¹ Suresh et al.⁷² studied the turning of hardened AISI 4340 steel by employing the Taguchi approach. They observed that cutting speed has the main effect on machining power. The results of this study established correlations between cutting parameters and machining force, power, specific cutting force, tool wear and surface roughness.

Mativenga and Rajemi tried to determine the optimum tool life to achieve a minimum energy footprint during turning operations of EN8 (AISI 1040) steel billets and found optimal values for depth of cut, feed rate and cutting speed. According to this study, at high values for feed rate and depth of cut, specific energy consumption is reduced.³⁴ Gupta et al.⁷³ introduced a Taguchi–fuzzy approach to optimize machining parameters while optimizing surface roughness, tool life, cutting force and power consumption using the model proposed by Aggarwal et al.⁷⁴

RSM is a set of mathematical and actuarial techniques used to model and analyse problems in which a response of interest is influenced by several parameters and the objective is to optimize this response.⁶² RSM was introduced by Box and Wilson⁷⁵ and later popularized by Montgomery.⁶² Abhang and Hameedullah developed a model to predict power consumption during a turning operation of EN-31 Steel using RSM. They used ANOVA for data analysis and RSM to develop the first-order and second-order models in order to predict power consumption. They were able to calculate optimum values for cutting speed, feed rate, depth of cut and nose radius in order to minimize power consumption.⁷⁶ Campatelli employed RSM to optimize cutting parameters in order to minimize power consumption during a milling operation of carbon steel.⁷⁷ Bhushan outlined experimental studies to investigate the effects of cutting speed, feed rate, depth of cut and nose radius on CNC turning of 7075 Al alloy SiC composite. The optimized cutting parameters were obtained in order to achieve minimum power consumption and maximum tool life using RSM.⁷⁸

Some scholars have employed both methods and compared results. Yan et al. applied weighted grey relational analysis and RSM to perform a multi-objective optimization. The objectives were to concurrently optimize surface roughness, material removal rate and cutting energy. The Taguchi methodology was also employed in order to confirm the developed method.⁷⁹

Kadirgama and Abou-El-Hossein⁸⁰ also used RSM to create models that predict power consumption. They built both first- and second-order models and compared the results. They concluded that the second-order model was more accurate than first-order model. The model was later improved by adopting factorial design to determine the effect of power on cutting force.⁸¹

Aggarwal et al. used both RSM and Taguchi techniques and presented a comparative analysis in order to optimize power consumption during a CNC turning operation of AISI P-20 tool steel in different cutting environments. The experiments were designed using an orthogonal array and face-centred central composite. Although the results of both methods were similar, it took nearly twice as long to use RSM than the Taguchi method to design the experiments. Also, the model developed using RSM is with respect to parameters, their interactions and square terms, but the model designed using the Taguchi method investigates only three interactions. Moreover, the RSM three-dimensional (3D) surface provides a better visualization of parameter effects.⁷⁴ Jou et al. integrated the Taguchi method and RSM in order to optimize process parameters for injection moulding. In this work, the Taguchi method was employed to obtain significant parameters and RSM was employed to develop a model for optimization.⁸² A summary of power optimization studies using DOE methods is provided in Table 5.

Table 5.

Summary of energy optimization studies using DOE methods.

Author	Work piece material	Cutting tool	Cutting parameters						Method
Author	Work piece material	Cutting tool	Cutting speed	Feed rate	Depth of cut	Width of cut	Nose radius	Cutting fluid	Method
Thomas et al.⁶³	ANSI 1018 carbon steel	Cemented carbide	✓	✓	✓				FFD
Yang and Tarng⁶⁴	S45C steel	Tungsten carbide	✓	✓	✓				Taguchi
Kadirgama and Abou-El-Hossein⁸⁰	618 stainless steel	Coated carbide	✓	✓	✓	✓			RSM
Kadirgama and Abou-El-Hossein⁸¹	618 stainless steel	Coated carbide	✓	✓	✓	✓			RSM + factorial design
Aggarwal et al.⁷⁴	AISI P-20 tool steel	TiN-coated tungsten carbide	✓	✓	✓		✓		RSM + Taguchi
Bhattacharya et al.⁶⁹	AISI 1045	Coated carbide	✓	✓	✓				Taguchi
Abhang and Hameedullah⁷⁶	EN-31 steel	Tungsten carbide	✓	✓	✓		✓		RSM
Fratila and Caizar⁶⁵	AlMg3 (EN AW 5754)	HSS	✓	✓	✓			✓	Taguchi
Gupta et al.⁷³	AISI P-20 tool steel	TiN-coated tungsten carbide	✓	✓	✓		✓		Taguchi + fuzzy
Babu et al.⁶⁷	Aluminium	Carbide	✓	✓	✓				Taguchi
Hanafi et al.⁷⁰	PEEK-CF30	TiN coated	✓	✓	✓				Taguchi
Bhushan⁷⁸	7075 Al alloy 15	Tungsten carbide	✓	✓	✓		✓		RSM
Yan and Li⁷⁹	Carbon steel	Carbide	✓	✓	✓	✓			RSM + Taguchi
Camposeco-Negrete⁷¹	AISI 6061 T6	Silicon carbide	✓	✓	✓				Taguchi
Adinarayana et al.⁶⁶	4340 alloy steel	Coated carbide	✓	✓	✓				Taguchi
Kulkarni et al.⁶⁸	AISI 1040 steel	Coated carbide	✓	✓	✓		✓	✓	Taguchi
Campatelli et al.⁷⁷	AISI 1050 carbon steel	Carbide	✓	✓	✓	✓			RSM

RSM: response surface methodology; HSS: high-speed steel; FFD: Full Factorial Design.

Artificial intelligence methods

As mentioned previously, cutting parameters are important factors for increasing the efficiency and quality of machining processes. In addition to empirical methodologies, artificial intelligence techniques have been used to select optimum cutting parameters in order to achieve more efficiency in cutting processes.

As the number of constraints increases the complexity of the machining, the size of the optimization problem increases as well. Although specific problems can be solved with conventional optimization techniques, they tend to yield local optimal solutions. Therefore, optimization problems typically are solved using non-traditional optimization techniques. Dynamic programming is one method for solving this class of optimization problem, which mainly involves simultaneously determining the optimum number of passes and depth of cut.^83–86 A dynamic programming model for multi-pass turning operations under constraints of force, cutting power and surface finish has been developed by Walvekar and Lambert⁸⁷ In addition, cognitive paradigms such as artificial neural networks (ANNs), simulated annealing (SA), genetic algorithms (GAs), ant colony approaches, bee colony approaches and nonlinear programming are included in modern heuristic methods. Despite being appropriate, some of these methods are slow in finding the optimal solution. Saravanan et al.’s⁸⁸ results show that non-conventional and conventional optimization techniques yield results that are 1%–3% different. Among the mentioned artificial intelligence techniques, SA, GAs and ANNs are used the most to optimize machining parameters. Important studies using these techniques are discussed in the following sections.

SA

SA is a nature-inspired method adapted from gradual cooling process of metals in nature. SA has been used in numerous studies in order to obtain optimal cutting parameters with different objectives, where the cutting force or cutting power is used as an optimization constraint. The main advantage of this method is that unlike traditional methods, there is no need to determine gradient descent, so it can be used for all types of objective and constraint functions. Moreover, a global minimum can be achieved using SA instead of a local minimum.⁸⁹ A limitation of this method is a high likelihood of revisiting recent solutions, thus requiring more iterations and more time to find the best solution. Some important studies using this method are summarized in Table 6, which also presents related objectives and constraints (additional to cutting parameters).

Table 6.

A summary of studies using GA and SA.

Author	Objective	Constraint	Process	Method	Remark
Wang et al.⁹⁰	Production time	Power consumption	Milling	SA + GA	SA and GA are used to optimize the cutting parameters during a multi-pass milling process
		Arbor strength
		Arbor deflection
Khan et al.⁹¹	Production cost	Cutting force	Turning	SA + GA	SA and GA are used to present several nonlinear and non-convex machining models in order to obtain optimum cutting parameters during a turning operation
		Cutting power
		Tool life
		Temperature
		Limitations on the value of the depth of cut in removing A in n passes
Kolahan and Abachizadeh⁹²	Total cost	Tool life	Turning	SA	An SA algorithm was developed in order to optimize cutting parameters during turning operation of a cylindrical workpiece
		Cutting force
		Cutting power
Sardinas et al.⁹³	Tool lifeOperation time	Cutting force	Turning	GA	GA was used in a multi-objective optimization problem to obtain optimal cutting parameters for turning processes
		Cutting power
		Surface roughness
Cus et al.⁹⁴	Production rateOperation costSurface roughness	Cutting power	Milling	GA	GA was used to optimize cutting parameters for ball-end milling process
Cus et al.⁹⁴	Production rateOperation costSurface roughness	Cutting force	Milling	GA

SA: simulated annealing; GA: genetic algorithm.

GA

Compared to traditional optimization paradigms, GA, which is based on the principles of natural biological evolution, is robust and global. Since a GA is applicable without reference to domain-specific heuristics, it is used extensively in machine learning, function optimizing and system modelling.

The advantages of this method are that it can be applied to complex objective functions with both discrete and continuous variables⁹⁵ and can automatically search for nonlinear connections between the inputs and outputs.⁹⁶ The main limitations of a GA are that convergence is not guaranteed and it may need a long execution time to create optimal outputs.⁹⁷ Some important studies using GA are also summarized in Table 6.

ANNs

ANNs are a family of statistical learning models used to optimize cutting parameters that are inspired by biological neural networks. The main advantages of ANNs are requiring less formal statistical training, ability to implicitly detect nonlinear relationships between dependent and independent variables and also ability to detect all possible interactions between predictor variables. The disadvantages are greater computational burden and proneness to overfitting.

In the reduced model outlined by Borgia et al.,⁹⁸ the analysis of the energy consumption of a machine tool during face milling operations is characterized by a minimum set of significant parameters describing the product, the process and the machine. A feed forward neural network represents the influence of these parameters with 20 inputs, 2 hidden layers and 1 output. Kant and Sangwan⁹⁹ used an ANN and support vector regression modelling techniques to predict the power consumed during a machining process. A real machining experiment was performed to evaluate the capability of these techniques to predict the amount of power consumed. The Taguchi method was used to study the effect of all the parameters with a minimum possible number of experiments.

The influence of process parameters on power consumption was analysed by Quintana et al.¹⁰⁰ in a series of 300 experiments on a three-axis vertical milling centre used for high-speed ball-end milling operations with AISI H13 steel. The data collected from the energy consumption of the spindle and the x-, y- and z-axes of the machine tool were used to develop an ANN model. The model accurately predicts power consumption during operations. Yun-qi et al.¹⁰¹ suggested a knowledge-based method in order to predict energy consumption using an ANN and performed verification studies as well. A chronological distribution of all optimization-related articles is provided in Figure 10, and the most important articles related to power optimization techniques are provided in Table 7, ranked based on the number of citations.

Figure 10.

Chronological distribution of optimization articles.

Table 7.

Most cited DOE articles.

Rank	Reference	Journal	Method	No. of citations
1	Box and Wilson⁷⁵	Journal of the Royal Statistical Society	RSM	3778
2	Yang and Tarng⁶⁴	Journal of Materials Processing Technology	Taguchi	708
3	Sardinas et al.⁹³	Engineering Applications of Artificial Intelligence	GA	168
4	Aggarwal et al.⁷⁴	Journal of Materials Processing Technology	RSM + Taguchi	101
5	Bhattacharya et al.⁶⁹	Journal of Production Engineering	Taguchi	84
6	Wang et al.⁹⁰	International Journal of Machine Tools & Manufacture	SA + GA	82
7	Khan et al.⁹¹	Computers Operations Research	SA + GA	80
8	Fratila and Caizar⁶⁵	Journal of Cleaner Production	Taguchi	64
9	Bhushan⁷⁸	Journal of Cleaner Production	RSM	53
10	Cus et al.⁹⁴	Journal of Materials Processing Technology	GA	44

RSM: response surface methodology; GA: genetic algorithm; SA: simulated annealing.

Conclusion

In this article, a comprehensive review of techniques for increasing the energy efficiency of machining systems has been presented. The distribution of articles with respect to methods is shown in Figure 11. Methodologies for modelling and analysing the energy consumption of machine tools in order to improve the energy efficiency of machining processes have been reviewed in-depth. The results of these studies reveal that energy consumption modelling, even for specific machine tools or machining processes, can be used to significantly increase energy efficiency. Many characterization studies conclude that auxiliary components of machine tools consume a lot more energy than the energy specifically used during cutting. This fact demands more studies to reduce idling times of machine tools by holistically analysing process chains. Furthermore, some studies point out that technical building services (TBS) in a factory building such as heating, cooling and lighting consume even more energy than all the machine tools require. Hence, these services should be worked on to provide more system efficiency.

Figure 11.

Distribution of articles with respect to methods.

Furthermore, optimization techniques aimed at minimizing power consumption at the process level during machining operations have been reviewed. In many studies, cutting parameters were identified as one of the main factors affecting the amount of consumed energy. Also, studies focused on obtaining optimal cutting parameters using different optimization techniques have been covered in this review. The literature survey revealed that improving the energy efficiency of machine tools is important. More research is required to develop a machining system to directly improve real-time energy optimization beyond optimizing time and cost.

In recent years, more attention has been paid to energy efficiency issues in manufacturing systems. In order to reduce energy consumption, it is necessary to develop energy-efficient techniques by predicting the behaviour and performance of machining systems, optimizing mechanical arrangements and selecting optimal cutting parameters. Major technological barriers have been identified by analysing limitations associated with the existing approaches. Beyond identifying challenges to energy efficiency in machining operations, major scientific and technological advancements in this area have been consolidated and presented.

Future work

Many opportunities for future research exist, given the diverse array of machining processes and machine tools. Today, the machining processes used in many industries such as automotive manufacturing, defence/aerospace and energy are very advanced and encompass many different auxiliary components to ensure both process stability and increased throughput. It should be noted that even though increased throughput for a machine tool–based process chain would improve overall energy efficiency, this will require more powerful auxiliary equipment which can demand even more energy, which will defeat its purpose. Nevertheless, it is possible to extend research related to machining processes and systems in the following directions:

The main limitation of most of the existing works is that they are applicable to specific machine tools, workpiece materials, cutting or machining processes. This limitation is as an important barrier to widespread adoption of techniques developed in academia. The effective use of material property libraries during the development of methods and systematic tools to characterize the energy efficiency of machine tools could help overcome this limitation.

The objectives of the existing cutting parameter optimization methods are typically cost, time and production rate. It would be beneficial to further optimize cutting parameters using artificial intelligence to determine optimal cutting conditions for minimizing the energy consumption of machining processes considering quality factors as constraints of optimization.

To further improve the energy efficiency of machining operations, comprehensive analyses of energy consumption data during processes should be performed. Compiling a complete database of information about machine tools, cutting tools, materials and energy sources will facilitate more accurate energy optimization.

Energy prediction models have mostly been developed for basic processes and simple machine tools such as three-axis milling machines or lathes. New models for more complicated machining processes such as turn-milling and five-axis machining would also be useful as advanced machining is used in many industries (e.g. automotive, aerospace). Further investigation of advanced processes other than machining, such as micro-wire electrical discharge machining (WEDM), friction stir-welding, and chemical vapour deposition, would be useful for reducing energy consumption in various industries.

Footnotes

Acknowledgements

The authors are sincerely grateful for the continued support of OSTIM Industrial Park management.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

This study was funded by SAN-TEZ project No. 00979.stz.2011-12 of the Turkish Ministry of Science, Technology and Industry.

References

EIA. Annual energy review, 2010, http://large.stanford.edu/courses/2012/ph241/druzgalski2/docs/aer.pdf

IEA. Worldwide trends in energy use and efficiency: key insights from IEA indicator analysis, 2008, https://www.iea.org/publications/freepublications/publication/Indicators_2008.pdf

Diaz

Dornfeld

Horvath

A comparative analysis of the environmental impacts of machine tool manufacturing facilities. J Clean Prod 2015; 95: 223–231.

Avram

Stroud

Xirouchakis

A multi-criteria decision method for sustainability assessment of the use phase of machine tool systems. Int J Adv Manuf Tech 2011; 53: 811–828.

Duflou

Sutherland

Dornfeld

. Towards energy and resource efficient manufacturing: a processes and systems approach. CIRP Ann: Manuf Techn 2012; 61: 587–609.

Fang

Uhan

Zhao

. A new shop scheduling approach in support of sustainable manufacturing. In: Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp. 305–310.

Pechmann

Schöler

Optimizing energy costs by intelligent production scheduling. In: In: Glocalized Solutions for Sustainability in Manufacturing, Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp.293–298.

Anderberg

Kara

Beno

Impact of energy efficiency on computer numerically controlled machining. Proc IMechE, Part B: J Engineering Manufacture 2010; 224(4): 531–541.

Kara

Bogdanski

Electricity metering and monitoring in manufacturing systems. In: Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp.1–10.

10.

Vijayaraghavan

Dornfeld

Automated energy monitoring of machine tools. CIRP Ann: Manuf Techn 2010; 59(1): 21–24.

11.

Pang

Gan

. Classification of energy consumption patterns for energy audit and machine scheduling in industrial manufacturing systems. Trans Inst Meas Control 2012; 35(5): 583–592.

12.

Avram

Xirouchakis

Evaluating the use phase energy requirements of a machine tool system. J Clean Prod 2011; 19(6–7): 699–711.

13.

Yang

Park

The prediction of cutting force in ball-end milling. Int J Mach Tool Manu 1991; 31(1): 45–54.

14.

Fang

Jawahir

IS.

Analytical predictions and experimental validation of cutting force ratio, chip thickness, and chip back-flow angle in restricted contact machining using the universal slip-line model. Int J Mach Tool Manu 2002; 42(6): 681–694.

15.

Yen

Söhner

Lilly

. Estimation of tool wear in orthogonal cutting using the finite element analysis. J Mater Process Tech 2004; 146(1): 82–91.

16.

Kishawy

Kannan

Balazinski

An energy based analytical force model for orthogonal cutting of metal matrix composites. CIRP Ann: Manuf Techn 2004; 53(1): 91–94.

17.

Palanisamy

Rajendran

Shanmugasundaram

. Prediction of cutting force and temperature rise in the end-milling operation. Proc IMechE, Part B: J Engineering Manufacture 2006; 220(10): 1577–1587.

18.

Huang

Liang

SY.

Modeling of cutting forces under hard turning conditions considering tool wear effect. J Manuf Sci Eng 2005; 127(2): 262–270.

19.

Bhattacharyya

Sengupta

Mukhopadhyay

Cutting force-based real-time estimation of tool wear in face milling using a combination of signal processing techniques. Mech Syst Signal Pr 2007; 21(6): 2665–2683.

20.

Sharma

AK.

Cutting tool wear estimation for turning. J Intell Manuf 2008; 19(1): 99–108.

21.

Astakhov

Xiao

A methodology for practical cutting force evaluation. J Mech Sci Technol 2008; 12(3): 325–347.

22.

Jerard

Fussell

BK.

Energy based cutting force model calibration for milling. Comput Aided Des Appl 2007; 4(1–4): 341–351.

23.

Cuppini

D’errico

Rutelli

Tool wear monitoring based on cutting power measurement. Wear 1990; 139(2): 303–311.

24.

Shao

Wang

Zhao

XM.

A cutting power model for tool wear monitoring in milling. Int J Mach Tool Manu 2004; 44: 1503–1509.

25.

Saidur

A review on electrical motors energy use and energy savings. Renew Sust Energ Rev 2010; 14: 877–898.

26.

Abele

Sielaff

Schiffler

. Analyzing energy consumption of machine tool spindle units and identification of potential for improvements of efficiency. In: Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp.280–285.

27.

De Filippi

Ippolito

Micheletti

. NC machine tools as electric energy users. CIRP Ann: Manuf Techn 1981; 30(1): 323–326.

28.

Kordonowy

DN.

A power assessment of machining tools. BSc Thesis, Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, 2001.

29.

Gutowski

Dahmus

Thiriez

. Electrical energy requirements for manufacturing processes. In: 13th CIRP international conference on life cycle engineering, Leuven, 31 May–2 June 2006, pp.623–627.

30.

Dahmus

Gutowski

. An environmental analysis of machining. In: ASME 2004 international mechanical engineering congress and exposition, Anaheim, CA, 13–19 November 2004, pp.643–652. New York: ASME.

31.

Gutowski

Murphy

Allen

. Environmentally benign manufacturing: observations from Japan, Europe and the United States. J Clean Prod 2005; 13(1): 1–17.

32.

Devoldere

Dewulf

Deprez

. Improvement potential for energy consumption in discrete part production machines. In: Advances in life cycle engineering for sustainable manufacturing businesses-14th CIRP international conference on life cycle engineering, Waseda University, Tokyo, Japan, 11–13 June 2007, pp.311–316. London: Springer.

33.

Rajemi

Mativenga

Aramcharoen

Sustainable machining: selection of optimum turning conditions based on minimum energy considerations. J Clean Prod 2010; 18(10–11): 1059–1065.

34.

Mativenga

Rajemi

MF.

Calculation of optimum cutting parameters based on minimum energy footprint. CIRP Ann: Manuf Techn 2011; 60: 149–152.

35.

Schischke

Hohwieler

Feitscher

. Energy-using product group analysis – lot 5. Berlin, 2012, http://www.evpg.bam.de/de/ebpg_medien/entr5/405_studyf_12_08.pdf

36.

Zein

Kara

. An investigation into fixed energy consumption of machine tools. In: Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp.268–273.

37.

Liu

. Analysis and estimation of energy consumption for numerical control machining. Proc IMechE, Part B: J Engineering Manufacture 2012; 226(2): 255–266.

38.

Salonitis

Ball

Energy efficient manufacturing from machine tools to manufacturing systems. Proced CIRP 2013; 7: 634–639.

39.

Calvanese

Albertelli

Matta

. Analysis of energy consumption in CNC machining centers and determination of optimal cutting conditions. In: Re-engineering Manufacturing for Sustainability Proceedings of the 20th CIRP International Conference on Life Cycle Engineering, Singapore 17–19 April, 2013, pp.141–146, Springer

40.

Aramcharoen

Mativenga

PT.

Critical factors in energy demand modelling for CNC milling and impact of toolpath strategy. J Clean Prod 2014; 78: 63–74.

41.

Uluer

Unver

Akkuş

. A model for predicting theoretical process energy consumption of rotational parts using STEP AP224 features. In: Proceedings of the 20th CIRP international conference on life cycle engineering, Singapore, 17–19 April 2013, pp.141–146.

42.

Mori

Fujishima

Inamasu

. A study on energy efficiency improvement for machine tools. CIRP Ann: Manuf Techn 2011; 60(1): 145–148.

43.

Diaz

Redelsheimer

Dornfeld

. Energy consumption characterization and reduction strategies for milling machine tool use. In: Proceedings of the 18th CIRP international conference on life cycle engineering, Technische Universität Braunschweig, Braunschweig, Germany, 2–4 May 2011, pp.263–267.

44.

Diaz

Helu

Jarvis

. Strategies for minimum energy operation for precision machining. In: Proceedings of the MTTRF 2009 annual meeting, Shanghai, China, 8–9 July 2009.

45.

Kong

Choi

Yasui

. Software-based tool path evaluation for environmental sustainability. J Manuf Syst 2011; 30(4): 241–247.

46.

Oliveira

JFG

Ferraz

Jr Coelho

. Architecture for machining process and production monitoring based in open computer numerical control. Proc IMechE, Part B: J Engineering Manufacture 2008; 222(12): 1605–1612.

47.

Christoffersen

Larsen

Togeby

Empirical analysis of energy management in Danish industry. J Clean Prod 2006; 14: 516–526.

48.

Neugebauer

Wabner

Rentzsch

. Structure principles of energy efficient machine tools. CIRP J Manuf Sci Technol 2011; 4(2): 136–147.

49.

Liu

. An on-line approach for energy efficiency monitoring of machine tools. J Clean Prod 2012; 27: 133–140.

50.

Liu

. Characteristics of additional load losses of spindle system of machine tools. J Adv Mech Design Syst Manuf 2010; 4(7): 1221–1233.

51.

Balogun

Mativenga

PT.

Modelling of direct energy requirements in mechanical machining processes. J Clean Prod 2013; 41: 179–186.

52.

Kellens

Renaldi Dewulf

Duflou

. Knowledge Collaboration & Learning for Sustainable Innovation In PROCEEDINGS ERSCP-EMSU conference Delft, The Netherlands, 25–29 October 2010, pp. 1–13.

53.

Draganescu

Gheorghe

Doicin

CV.

Models of machine tool efficiency and specific consumed energy. J Mater Process Tech 2003; 141(1): 9–15.

54.

Kara

Unit process energy consumption models for material removal processes. CIRP Ann: Manuf Techn 2011; 60(1): 37–40.

55.

Guo

Loenders

Duflou

. Optimization of energy consumption and surface quality in finish turning. Proced CIRP 2012; 1: 512–517.

56.

Yoon

Moon

Pham

. Control of machining parameters for energy and cost savings in micro-scale drilling of PCBs. J Clean Prod 2013; 54: 41–48.

57.

Yoon

Lee

Kim

. Empirical power-consumption model for material removal in three axis milling. J Clean Prod 2014; 78: 54–62.

58.

Pervaiz

Deiab

Rashid

Prediction of energy consumption and environmental implications for turning operation using finite element analysis. Proc IMechE Part B: J Engineering Manufacture. Epub ahead of print 25 July 2014. DOI: 10.1177/0954405414541105.

59.

Dietmair

Verl

. Energy consumption modeling and optimization for production machines. In: Proceedings of the IEEE international conference on sustainable energy technologies, Singapore, 24–27 November 2008, pp.574–579. New York: IEEE.

60.

Dietmair

Verl

A generic energy consumption model for decision making and energy efficiency optimisation in manufacturing. Int J Sust Eng 2009; 2(2): 123–133.

61.

Larek

Brinksmeier

Meyer

. A discrete-event simulation approach to predict power consumption in machining processes. J Prod Eng 2011; 5: 575–579.

62.

Montgomery

DC.

Design and analysis of experiments. New York: Wiley, 1997.

63.

Thomas

Beauchamp

Youssef

. An experimental design for surface roughness and built-up edge formation in lathe dry turning. Int J Qual Sci 1997; 2(3): 167–180.

64.

Yang

Tarng

YS.

Design optimization of cutting parameters for turning operations based on the Taguchi method. J Mater Process Tech 1998; 84(1–3): 122–129.

65.

Fratila

Caizar

Application of Taguchi method to selection of optimal lubrication and cutting conditions in face milling of AlMg₃ . J Clean Prod 2011; 19: 640–645.

66.

Adinarayana

Prasanthi

Krishnaiah

Parametric analysis and multi objective optimization of cutting parameters in turning operation of AISI 4340 alloy steel with CVD cutting tool. Int J Res Eng Tech 2014; 3(2): 449–456.

67.

Babu

Kumar

Vishnuu

Optimization of cutting parameters for CNC turned parts using Taguchi’s technique. Int J Eng 2012; 3: 493–496.

68.

Kulkarni

Mandave

Sabnis

Optimization of power consumption for CNC turning Of AISI 1040 steel using Taguchi approach. Int J Innov Res Sci Eng Technol 2014; 3(8): 15383–15390.

69.

Bhattacharya

Das

Majumder

. Estimating the effect of cutting parameters on surface finish and power consumption during high speed machining of AISI 1045 steel using Taguchi design and ANOVA. Prod Eng 2009; 3: 31–40.

70.

Hanafi

Khamlichi

Cabrera

. Optimization of cutting conditions for sustainable machining of PEEK-CF30 using TiN tools. J Clean Prod 2012; 33: 1–9.

71.

Camposeco-Negrete

Optimization of cutting parameters for minimizing energy consumption in turning of AISI 6061 T6 using Taguchi methodology and ANOVA. J Clean Prod 2013; 53: 195–203.

72.

Suresh

Basavarajappa

Samuel

GL.

Some studies on hard turning of AISI 4340 steel using multilayer coated carbide tool. Measurement 2012; 45(7): 1872–1884.

73.

Gupta

Singh

Aggarwal

Taguchi-fuzzy multi output optimization (MOO) in high speed CNC turning of AISI P-20 tool steel. Expert Syst Appl 2011; 38(6): 6822–6828.

74.

Aggarwal

Singh

Kumar

. Optimizing power consumption for CNC turned parts using response surface methodology and Taguchi’s technique – a comparative analysis. J Mater Process Tech 2008; 200(1–3): 373–384.

75.

Box

GEP

Wilson

. On the experimental attainment of optimum conditions. J R Stat Soc 1951; 13(1): 1–45.

76.

Abhang

Hameedullah

Power prediction model for turning EN-31 steel using response surface methodology. J Eng Sci Technol Rev 2010; 3(1): 116–122.

77.

Campatelli

Lorenzini

Scippa

Optimization of process parameters using a response surface method for minimizing power consumption in the milling of carbon steel. J Clean Prod 2014; 66: 309–316.

78.

Bhushan

KR.

Optimization of cutting parameters for minimizing power consumption and maximizing tool life during machining of Al alloy SiC particle composites. J Clean Prod 2013; 39: 242–254.

79.

Yan

Multi-objective optimization of milling parameters e the trade-offs between energy, production rate and cutting quality. J Clean Prod 2013; 52: 462–471.

80.

Kadirgama

Abou-El-Hossein

KA.

Power prediction model for milling 618 stainless steel using response surface methodology. Am J Appl Sci 2005; 2(7): 1182–1187.

81.

Kadirgama

Abou-El-Hossein

Torque . Power and cutting force prediction model by using response surface method and factorial design. Eur J Sci Res 2007; 18(1): 20–44.

82.

Jou

Lin

Lee

. Integrating the Taguchi method and response surface methodology for process parameter optimization of the injection molding. Int J Appl Math Inf Sci 2014; 8(3): 1277–1285.

83.

Iwata

Murotsu

Iwatsubo

. A probabilistic approach to the determination of the optimum cutting conditions. J Eng Ind 1972; 94(4): 1099–1107.

84.

Agapiou

The optimization of machining operations based on a combined criterion, part-1: the use of combined objectives in single pass operations. J Eng Ind: T ASME 1992; 114(4): 500–507.

85.

Agapiou

The optimization of machining operations based on a combined criterion, part-2: multipass operations. J Eng Ind: T ASME 1992; 114(4): 508–513.

86.

Shin

Joob

YS.

Optimization of machining conditions with practical constraints. Int J Prod Res 1992; 30(12): 2907–2919.

87.

Walvekar

Lambert

BK.

An application of geometric programming to machining variable selection. Int J Prod Res 1970; 8(3): 241–245.

88.

Saravanan

Asokan

Sachithanandam

Comparative analysis of conventional and non-conventional optimisation techniques for CNC turning process. Int J Adv Manuf Technol 2001; 17: 471–476.

89.

Tarng

Chung

LK.

Determination of optimal cutting parameters in wire electrical discharge machining. Int J Mach Tool Manu 1995; 35(12): 1693–1701.

90.

Wang

Rahman

Wong

. Optimization of multi-pass milling using parallel genetic algorithm and parallel genetic simulated annealing. Int J Mach Tool Manu 2005; 45: 1726–1734.

91.

Khan

Prasad

Singh

Machining condition optimization by genetic algorithms and simulated annealing. Comput Oper Res 1997; 24(7): 647–657.

92.

Kolahan

Abachizadeh

Optimizing turning parameters for cylindrical parts using simulated annealing method. Proc World Acad Sci E 2008; 2: 10–20.

93.

Sardinas

Santana

Brindis

EA.

Genetic algorithm-based multi-objective optimization of cutting parameters in turning processes. Eng Appl Artif Intell 2006; 19(2): 127–133.

94.

Cus

Milfelner

Balic

An intelligent system for monitoring and optimization of ball-end milling process. J Mater Process Tech 2006; 175(1–3): 90–97.

95.

Manolas

Gialamas

Frangopoulos

. A genetic algorithm for operation optimization of an industrial cogeneration system. Comput Chem Eng 1996; 20(2): 1107–1112.

96.

Cus

Balic

Optimization of cutting process by GA approach. Robot Comp Int Manuf 2003; 19: 113–121.

97.

Mukherjee

Ray

PK.

A review of optimization techniques in metal cutting processes. Comput Ind Eng 2006; 50(1–2): 15–34.

98.

Borgia

Pellegrinelli

Bianchi

. A reduced model for energy consumption analysis in milling. Proced CIRP 2014; 17: 529–534.

99.

Kant

Sangwan

KS.

Prediction and optimization of machining parameters for minimizing power consumption and surface roughness in machining. J Clean Prod 2014; 83: 151–164.

100.

Quintana

Ciurana

Ribatallada

Modelling power consumption in ball-end milling operations. Mater Manuf Processes 2011; 26(5): 746–756.

101.

Yun-qi

Min

Gang

. Knowledge-based calculation and prediction of energy consumption in productmanufacturing process. J South China Univ Technol 2009; 37(2): 14–19.

Energy efficiency of machining operations: A review

Abstract

Keywords

Introduction

Motivation for review

Research methodology

Categorical literature review

Energy consumption models

Theoretical modelling

Empirical modelling

Optimization of cutting parameters

Design of experiments

Artificial intelligence methods

SA

GA

ANNs

Conclusion

Future work

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

References