1 Introduction

Manufacturing is defined as the transformation of raw materials into products, being among the top human activities worldwide [1]. In general, human activities are responsible for increasing concentrations of greenhouse gases (GHG) which are linked to the rapid rise of global temperature. According to literature [2], the largest share of global greenhouses gases comes from the energy consumption (73.2%), with the main contributor being the industrial sector. The latest analysis from European Statistics (Eurostat) has placed manufacturing as the second highest contributor, second to energy sector that includes electricity, gas, steam, and air conditioning supply [3]. However, this definition of energy sector includes also demands from manufacturing sector. To address the urging issue of increasing carbon emissions, European Union (EU) has published directives for short-term sustainable growth [4] and long-term carbon neutral economy [5]. The short-term directives focus in decreasing the carbon emissions of products/services within EU, while the longer-term strategy aims to create a carbon neutral and circular economy in EU.

Carbon dioxide accounts for the largest share of greenhouse gases. In order to decrease the carbon emissions, the term of carbon footprint for all products and activities was introduced. The main definition of carbon footprint is the quantity of greenhouse gases expressed in terms of equivalent mass of carbon dioxide emitted by an individual, organization, process, product, or event within a specific boundary and is expressed in kgCO2-e [6]. There are standards and guidelines available for greenhouse gases and carbon footprint calculation, such as GHG protocol of World Resource Institute [7], ISO 14040 [8], and ISO 14064 [9].

Since industry is among the largest contributors of carbon dioxide emissions [10], several studies calculated the carbon emissions of processes, to investigate the order of magnitude of the carbon emissions [11, 12] to identify the carbon footprint-intensive steps [13, 14] and to optimize process for reduced carbon emissions [15,16,17,18,19]. According to the classification of Chryssolouris [1], the manufacturing processes can be either phase changing processes, which utilize not only electrical energy, but also other sources of energy such as natural gas, or non-phase changing processes, that primarily use electrical energy. The majority of carbon emissions is linked with the use of electrical energy and is therefore calculated by the product of energy consumed times a carbon emission factor. This carbon emission factor varies from country to country, or from area to area, since it is linked to the way electrical energy is generated [20]. A review and critical appraisal of the existing literature are missing, in order to identify the limitations of the current approaches, but also to formulate the methodology and highlight the importance of different components used in the calculation of carbon footprint for processes.

This review aims to investigate the environmental impact of manufacturing, by (1) collating the equations and approaches from literature, (2) formulating the general equations for carbon footprint calculation, regardless the process, and (3) identifying the impact of qualitative and quantitative parameters use the carbon footprint. The contents of this review are summarized in the following order: Section 2 includes the methodology for the performing of the literature review; Section 3 starts by defining key concepts of manufacturing processes and levels, followed by the main equations used for carbon footprint calculation at each level; Section 4 provides sensitivity analysis for three case studies (material removal, primary forming, and deforming); Section 5 includes the discussion, followed by the Section 6 and the conclusion of the work.

2 Methods

Literature search was conducted in Google scholar and Scopus to identify the papers to be included in this review with keywords: “process,” “manufacturing,” and “carbon emission calculation.” This search resulted in 146 papers, which were later reviewed to ensure their relevance to the scope of this review (Fig. 1).

Fig. 1
figure 1

Flowchart of present literature review, with categorization of papers in material removal, primary forming, and deforming processes

Full size image

Articles were included based on the following criteria:

  1. a)

    Type of process (material removal, primary forming, and deforming)

  2. b)

    Calculation of carbon emissions using mathematical equations

At the end, 79 papers were included and divided according to the process addressed:

The following information were extracted from the papers and can be found in the Appendix 1:

  1. a)

    Determination of the level of steps included in the carbon calculation (process level, machine level, system level)

  2. b)

    Type of process

3 General approach for carbon calculation in manufacturing

3.1 Definition of manufacturing processes and description of levels

According to Chryssolouris et al [93], manufacturing is defined “as the transformation of materials and information into goods” to cover customer needs. Besides this, manufacturing is accompanied with carbon emissions, with each product/process related to a specific carbon footprint, based on the specific material, machine tool, and energy used.

Manufacturing can be divided into three main levels: process, machine, and system, according to whether the reference is on the process itself, with or without the aid of auxiliary equipment and lastly, if other processes are included, which are either related to the factory/industry level or material and tool production, disposal, and transportation.

At each level, the inputs are energy, materials, and tools/equipment, and the outputs are products and carbon footprint of products (Fig. 2, Table 1). These parameters are classified into quantitative and qualitative parameters, depending on whether they are related to the source or emission factor or whether they are related to their value, respectively (Table 2).

Fig. 2
figure 2

Description of levels, with raw material, machine tool, and energy as inputs and product and carbon footprint as outputs

Full size image
Table 1 Description of each level and listing of components included
Full size table
Table 2 Quantitative and qualitative parameters, divided with regards to their relation to raw material, machine tool, and electrical energy relation, but also in terms of manufacturing level
Full size table

3.2 Carbon emissions at process level

Carbon emissions from the process are mainly related to the energy consumed during the process (Eprocess). The energy depends on the material and machine specifications and requirements. Energy can be generated from different sources and brought in different forms; however, in machine tools, the most frequent is electrical energy [1]. The carbon emission factor (CEFelec) depends on the source of power generation.

The carbon emissions of process are given by:

$$ {CE}_{process}={ CE F}_{ele\mathrm{c}}\cdotp {E}_{process} $$
(1)

CEFelec is related to the electrical energy generation source, while Eprocess is a quantitative parameter, namely related to how much energy is needed to complete this process. Eprocess depends on the raw material, the final product requirements, and the machine specifications, such as feed rate and process speed.

3.3 Carbon emissions at machine tool level

At the level of machine, the carbon emissions are given not only by the process, but also from the auxiliary equipment and consumption of tools and consumables:

$$ {CE}_{machine}={CE}_{process}+{CE}_{aux}+{CE}_{tool, cons}+\sum \limits_i{CE}_{i, cons} $$
(2)

where CEaux is the carbon emissions from auxiliary equipment (namely starting energy, basic energy, idle energy, and downtime energy), CEtool, cons carbon emissions from tool consumption, and CEi, cons carbon emissions from consumable consumption, where consumables can be lubricant, coolant, and chips.

Carbon emissions at machine level are also given by:

$$ {\displaystyle \begin{array}{c}{CE}_{machine}={ CE F}_{elec}\cdotp {E}_{process}+{ CE F}_{elec}\cdotp {E}_{aux}\\ {}+{ CE F}_{tool, cons}\cdotp {E}_{tool, cons}+\sum \limits_i{ CE F}_{i, cons}\cdotp {Q}_{i, cons}\end{array}} $$
(3)

where CEFi, cons is the carbon emission factor of tool consumption, Etool, cons energy required for tool consumption, CEFi, cons carbon emission factor of consumable consumption, and Qi, cons quantity of consumables consumed to complete the production of the final product.

Similarly, to the process level, some parameters are qualitative (related to the carbon emission factors) and quantitative (amount of energy, material and consumable), as seen in Table 2.

3.4 Carbon emissions at system level

At the system level, the carbon emissions are given by the sum of carbon emissions at machine level and carbon emissions from transport (CEi, transp), production (CEi, prod), and disposal (CEi, disp) of material, tools, and consumable:

$$ {CE}_{system}={CE}_{machine}+\sum \limits_i\left({CE}_{i, transp}+{CE}_{i, prod}+{CE}_{i, disp}\right) $$
(4)

In details, carbon emissions of the overall production are calculated by:

$$ {\displaystyle \begin{array}{c}{CE}_{system}={ CE F}_{elec}\cdotp {E}_{process}+{ CE F}_{elec}\cdotp {E}_{aux}\\ {}+{ CE F}_{tool, cons}\cdotp {E}_{tool, cons}+\sum \limits_i{ CE F}_{i, cons}\cdotp {Q}_{i, cons}\\ {}+\sum \limits_i\left[{ CE F}_{i, transp}\cdotp {Q}_{i, transp}+\left({ CE F}_{i, prod}\cdotp {Q}_{i, prod}+{ CE F}_{i, disp}\cdotp {Q}_{i, disp}\right)\frac{t}{T_{i, life}}\right]\end{array}} $$
(5)

where CEFelec is the carbon emission factor of electrical energy; Eprocess energy required for the process itself; Eaux energy required for the operation of auxiliary/peripheral equipment; CEFtool, cons carbon emission factor for tool consumption; Etool, cons energy required for tool consumption; CEFi, cons carbon emission factor for material and consumable consumption; Qi, cons amount of material and consumable used in the operation; CEFi, transp carbon emission factor during transport to and from the production site; Qi, transp amount of transported quantity; CEFi, prod carbon emission factor of production of material, tool, and consumables; Qi, prod quantity produced; CEFi, disp carbon emission factor of disposal of material, tool, and consumables; Qi, disp quantity that will be disposed or recycled; and Ti, life lifetime of material, tool, and consumables.

4 Sensitivity analysis

4.1 Material removal

Using data from literature [16], a sensitivity analysis on turning of AISI4140 using carbide inserts in wet conditions was performed.

For the process level, energy consumed during turning is calculated based on the material removal volume (Vc) and energy requirements (kC):

$$ {E}_c={k}_C\cdotp {V}_c $$
(6)

Carbon emissions due to electrical energy are given by the product of carbon emission factor for electrical energy (CEFelec) and the energy consumed for the turning (Ec):

$$ {CE}_c={ CE F}_{elec}\cdotp {E}_c $$
(7)

Energy is consumed when no turning takes place, but spindle is turning (Eidle) and also during starting up (Ebasic) and shutting down (Edowntime)of the machines, being in turn responsible for carbon emissions:

$$ {CE}_{basic}={ CE F}_{elec}\cdotp {E}_{basic} $$
(8)
$$ {CE}_{downtime}={ CE F}_{elec}\cdotp {E}_{downtime} $$
(9)
$$ {CE}_{idle}={ CE F}_{elec}\cdotp {E}_{idle} $$
(10)

During machine level, there are also the carbon emissions due to tool wear and consumption and due to tool mass (Mtool) and tool lifetime (Tlife):

$$ {CE}_{tool, cons}={ CE F}_{tool, cons}\cdotp {M}_{tool}\cdotp \frac{t}{T_{life}} $$
(11)

In addition, coolant consumption results in carbon emissions, based on the quantity of coolant consumed (Qcool):

$$ {CE}_{cool, cons}={ CE F}_{cool, cons}\cdotp {Q}_{cool}\cdotp t $$
(12)

At the system level, carbon emissions are related to the production and disposal of material, coolant, and waste. In this case, coolant disposal, material production, and material disposal are taken under consideration, based on the quantity produced (Qcool, ∆mmaterial) or disposed (Mchip):

$$ {CE}_{cool, disp}={ CE F}_{cool, disp}\cdotp {Q}_{cool}\cdotp t $$
(13)
$$ {CE}_{material, prod}={ CE F}_{material, prod}\cdotp \Delta {m}_{material} $$
(14)
$$ {CE}_{chip}={ CE F}_{chip, disp}\cdotp {M}_{chip} $$
(15)

Parameters in Equations (7)–(15) are characterized either as qualitative or quantitative, depending on whether they are linked to the way that power is generated, tool and coolant liquids are consumed, coolant and chips are disposed, and material is produced or linked to the quantity of energy, material, tool, and coolant consumed, produced, or disposed. Parameters are categorized in Table 3, while sensitivity factors are summarized in Table 4.

Table 3 Qualitative and quantitative parameters for carbon emission calculations, on each level (process, machine, and system) or factor (material, machine tool, electrical parameter)
Full size table
Table 4 Sensitivity coefficients from the sensitivity analysis
Full size table

The results from the sensitivity analysis are described in the following paragraphs, tables, and figures. In the case of carbon emission factor for electrical energy, the highest value is at the process level, showing that any change on the carbon emission factor for electrical energy will affect more the carbon emissions of process, much less the machine level with the least effect on the carbon emissions at the system level. Any variations of the rest of carbon emission factors will have no impact on the carbon emission at process level. Regarding the rest of carbon emission factors, machine level is affected by carbon emission factor of tool and coolant consumption, with the tool consumption being the second heaviest factor, with the strongest factor being the carbon emission factor for electrical energy. At the system level, the heaviest carbon emission factor is the one of material production, followed by the carbon emission factor of electrical energy and carbon emission factor of tool consumption. The impact of each carbon emission factor per level can be found in the sensitivity table (Table 4), and relevant plots are found in the Appendix 2.

Tornado plots are used in sensitivity analysis to show the impact of factors on the overall value. In Fig. 3, the only factor is the carbon emission factor of electrical energy, while in Fig. 4, three factors are affecting the carbon emissions, namely carbon emission factors for electrical energy, tool consumption, and coolant consumption. As seen in Fig. 4, the order of weight from heavier to lighter is carbon emission factor for electrical energy, carbon emission factor for tool consumption, and carbon emission factor for coolant consumption. In Fig. 5, the heaviest factor is the carbon emission factor of material production, followed by carbon emission factor of electrical energy. Carbon emission factors of tool consumption and chip disposal are also contributing to the carbon emissions at system level, with the weight of carbon emission factor of coolant disposal and coolant consumption being so negligible that are close to zero on Fig. 5.

Fig. 3
figure 3

Tornado plot for material removal at process level. Equation (7) was used for this plot (CEc = CEFelec ∙ Ec), with amount of energy consumed during turning (Ec) remaining constant, while carbon emission factor ranging from 0.39 to 1.17 kgCO2/kWh. The rest of the carbon emission factors are not involved at process level; hence, they do not have impact on the carbon emissions at process level

Full size image
Fig. 4
figure 4

Tornado plot for material removal at machine level (left), with zooming into the carbon emission factors of tool and coolant consumption at machine level (right). Equations (8)–(12) were used for this plot, with only carbon emission factors ranging, while the rest parameters remaining the same. The carbon emission factor for electrical energy (CEFelec )still is the heaviest contributing factor, followed by carbon emission factor of tool consumption (CEFtool, cons) and carbon emission factor of coolant consumption (CEFcool, cons)

Full size image
Fig. 5
figure 5

Tornado plot for material removal at system level (left), zooming into the carbon emission factors from heaviest contributing factor (CEFmaterial, prod) to the lightest contributing factor (CEFcool, disp). Equations (13)–(15) were used. Overall, the heaviest contributing factor is the carbon emission factor for material production, followed by the carbon emission factor for electrical energy

Full size image

4.2 Primary forming

To perform the sensitivity analysis for a primary forming process, the example from Liu et al. [20] was used, which was based on the casting of box-beams from base sand, dry clay, and water.

For the process level, energy was consumed during the pressing of the sand-clay-water mixture and the carbon footprint is given by:

$$ C{E}_{comp}=\frac{P_{comp}}{W_{c\mathrm{o} mp}}\cdotp {CEF}_{ele} $$
(16)

which was calculated based on the power of machine for the compression (Pcomp), rate of compression (Wcomp), and carbon emission factor of electrical energy (CEFele).

Electrical energy is consumed in other auxiliary process, such as mold drying (CEdry), shake out (CEsha), and cleaning (CEcle). Carbon emissions of the auxiliary processes are given by:

$$ {CE}_{dry}=\left[{U}_{san}^{dry}\cdotp {M}_{san}\left(1-{\eta}_{water}\right)\Delta T\right]{ CE F}_{ele}+\left({U}_{water}\cdotp \Delta {T}_{water}+Q\right){M}_{san}{\eta}_{\mathrm{w} ater}{ CE F}_{ele} $$
(17)
$$ {CE}_{sha}={P}_{sha}\cdotp \frac{M_{san}}{W_{sha}}{ CE F}_{ele} $$
(18)
$$ {CE}_{cle}={P}_{cle}\cdotp \frac{A_{cas}}{W_{cle}}{ CE F}_{ele} $$
(19)

where \( {U}_{san}^{dry} \) is the specific heat capacity of dried clay sand, Msan mass of casting sand, ηwater fraction of water, Uwater specific heat capacity of water, ΔTwater the difference between water’s boiling point and its initial temperature, Q evaporation heat, Psha power of the shakeout machine and Wsha shakeout rate, Pcle power of cleaning machine, Acas surface area of casting, and Wcle cleaning rate.

At the system level, carbon emissions come from the sand production (CEnew), sand reclamation (CErec), and sand mixing (CEmix):

$$ {CE}_{new}=\left({\eta}_{new}\cdotp {M}_{san}\right){ce}_{rec} $$
(20)
$$ {\displaystyle \begin{array}{c}{ce}_{new}={\eta}_{cla}\cdotp \left({E}_{gas}^{cla}\cdotp {CEF}_{gas}+{E}_{coa}^{cla}\cdotp {CEF}_{coa}+{E}_{cru}^{cla}\cdotp {CEF}_{cru}+{E}_{die}^{cla}\cdotp {CEF}_{die}+{E}_{ele}^{cla}\cdotp {CEF}_{ele}\right)\\ {}+{\eta}_{san}\cdotp \left({E}_{gas}^{san}\cdotp {CEF}_{gas}+{E}_{coa}^{san}\cdotp {CEF}_{coa}+{E}_{cru}^{san}\cdotp {CEF}_{cru}+{E}_{die}^{san}\cdotp {CEF}_{die}+{E}_{ele}^{san}\cdotp {CEF}_{ele}\right)\end{array}} $$
(21)
$$ {CE}_{rec}=\frac{\eta_{rec}\cdotp {M}_{san}}{W_{rec}}\cdotp {P}_{rec}\cdotp {CE F}_{ele} $$
(22)
$$ {CE}_{mix}=\frac{\eta_{rec}\cdotp {M}_{san}}{W_{rec}}\cdotp {P}_{rec}\cdotp {CE F}_{ele} $$
(23)

where ηnew is the mass fraction of new sand and Msan mass of casting sand;\( {E}_{gas}^{cla},{E}_{coa}^{cla},{E}_{cru}^{cla},{E}_{die}^{cla},\mathrm{and}\ {E}_{ele}^{cla} \) energy consumptions from clay and \( {E}_{gas}^{san},{E}_{coa}^{san},{E}_{cru}^{san},{E}_{die}^{san},\mathrm{and}\ {E}_{ele}^{san} \) energy consumptions from sand; Wrec reclamation rate and Prec power of reclamation; and Wmix the mixing rate and Pmix power of mixing equipment.

Finally, the carbon emissions due to transport of raw material from a distance of 271.07 km (average distance in China) were calculated by:

$$ {CE}_{transp}=\left({ CE F}_{ele}\cdotp {E}_{ele c, transp}+{ CE F}_{coa}\cdotp {E}_{coa, transp}+{ CE F}_{die}\cdotp {E}_{die, transp}\right)\cdotp d $$
(24)

Parameters in the carbon emission calculation can be divided per process, machine, and system level, while the parameters are related to electrical energy-, machine-, and material-related parameters. In each group, the parameters are either qualitative, in terms of footprint, or quantitative, in terms of how much energy or power or material is used (Table 5).

Table 5 Qualitative and quantitative parameters, related to electrical energy, machine, and material, in all levels of production (process, machine, system)
Full size table

The results from the sensitivity analysis can be found in the following table and figures. The values from sensitivity analysis can be found in Table 6 (with relevant spider charts in Appendix 2).

Table 6 Sensitivity coefficients from the sensitivity analysis
Full size table

Tornado charts were plotted using the sensitivity factors and their effect on carbon emissions at process, machine, and system levels. As seen in Fig. 6, the only contributing factor is carbon emission factor for electrical energy and so is in Fig. 7. Regarding system level, the heaviest contributing factor is the carbon emission factor of coal consumption, followed by the carbon emission factor for electrical energy and carbon emission factor for diesel consumption (Fig. 8). The carbon emissions factor of crude oil and gas consumption has negligible effect on carbon emissions at system level.

Fig. 6
figure 6

Tornado plot for primary forming at process level. The main contributing factor is carbon emission factor for electrical energy (CEFele) as determined by Equations 16. The rest of the carbon emission factors are at system level

Full size image
Fig. 7
figure 7

Tornado plot for primary forming at machine level. The main contributing factor is carbon emission factor for electrical energy (CEFele) as determined by Equations  16–19. The rest of the carbon emission factors are at system level

Full size image
Fig. 8
figure 8

Tornado plot for primary forming at system level (left), with zooming into the system level (top right and bottom right). The heaviest contributing factors are carbon emission factor of coal or carbon emission factor of diesel (top right), and the lightest contributing factors are carbon emission factor of gas or carbon emission factor of crude oil (bottom right). Equations 20–24 were used for this plot

Full size image

4.3 Deforming

The sensitivity analysis for a deforming process was performed for the incremental forming of roll-bonded Cu/Steel laminates from Al-Ghamdi and Hussain work [85].

The process was the pressing of the laminates with energy required given by:

$$ {E}_{pres}=\frac{W}{\eta_{Pres}} $$
(25)

W is the deforming work and ηPres efficiency of hydraulic press.

The carbon emissions of the process level are given by:

$$ {CE}_{pres}={E}_{pres}\cdotp {CE F}_{elec} $$
(26)

At the machine level, carbon dioxide is also emitted during lubricant consumption (CElub, cons), tool consumption (CEtool, cons), fixture consumption (CEfix, cons), annealing (CEanneal), bonding (CEbond), and cutting of the sheets:

$$ {CE}_{lub, cons}={t}_f\cdotp {CE F}_{lub, cons} $$
(27)
$$ {CE}_{tool, cons}=\frac{t_f}{T_{formingtool}}\cdotp {M}_{tool}\cdotp {CE F}_{tool, cons} $$
(28)
$$ {CE}_{fix, cons}=\frac{t_f}{T_{fix tures}}\cdotp {CE F}_{fix, cons} $$
(29)
$$ {CE}_{anneal}={E}_{anneal}\cdotp {CE F}_{elec} $$
(30)
$$ {CE}_{bond}={E}_{bond}\cdotp {CE F}_{elec} $$
(31)
$$ {CE}_{cut}={E}_{cut}\cdotp {CE F}_{elec} $$
(32)

where tf is the time required for forming, CEFlub, cons carbon emission factor for lubricant consumption, Tformingtool lifetime of forming tool, Mtool mass of tool, CEFtool, cons carbon emission factor for tool consumption., Tfixtures lifetime of fixtures, CEFfix, cons carbon emission factor for fixture usage, Eanneal energy consumption during annealing, Ebond energy consumption during bonding, and Ecut energy required for cutting of the sheets.

At the level of the system, the additional carbon emissions are given by the production of material (CEmat, prod) and synthetic lubricant (CElubr, prod):

$$ {CE}_{mat, prod}={M}_{Cu}\cdotp {CE F}_{Cu, prod}+{M}_{Steel}\cdotp {CE F}_{Steel, prod} $$
(33)
$$ {CE}_{lubr, prod}={V}_{lubr}\cdotp {CE F}_{lubr, prod} $$
(34)

where MCu is the mass of Cu used for the sheet, CEFCu, prod carbon emission factor for the production of Cu, MSteel mass of Steel used for the sheet, CEFSteel, prod carbon emission factor for the production of Steel, Vlubr volume of lubricant used during incremental forming, and CEFlubr, prod carbon emission factor for the production of lubricant.

Similarly, the parameters are grouped with regards to their link to electrical energy, machine, or material, in the level of carbon emission calculation (process, machine, system level). Each group is divided into qualitative (in terms of footprint) and quantitative (use of less energy, resources, and material) (Table 7).

Table 7 Qualitative and quantitative parameters, related to electrical energy, machine, and material, in all levels of production (process, machine, system)
Full size table

Sensitivity analysis was performed, and the results are the following. The sensitivity factors can be found in the relevant table (Table 8, with relevant spider charts in Figs. 12, 13, 14 in Appendix 2), showing that the strongest contributing factor at all levels is the carbon emission factor for electrical energy. Carbon emission factors of material production are also strong contributors at system level but not in the same degree as other processes because this specific process has some energy-intensive steps, such as annealing with duration of 2 h. The contribution of the rest of the carbon emission actors is negligible (<0.02).

Table 8 Sensitivity coefficients from the sensitivity analysis
Full size table

Tornado charts from the sensitivity analysis confirm that the strongest contributing factor at all levels is the carbon emission factor of electrical energy consumption (Figs. 9 and 11). At process level, as in both the previous examples, the only factor is the carbon emission factor of electrical energy (Fig. 9), while at machine level, there is a considerably small contribution from the carbon emission factor of the lubricant consumption during the deforming (Fig. 10). At system level, the second contributing factor is the carbon emission factor for steel production, closely followed by the carbon emission factor for Cu production (Fig. 11). The lightest contributing factor comes from the carbon emission factor of lubricant production.

Fig. 9
figure 9

Tornado plot for deforming at process level. Equation (26) (CEpres = Epres ∙ CEFelec) was used at process level, with only contributing factor being the carbon emission factor for electrical energy

Full size image
Fig. 10
figure 10

Tornado plot for deforming at machine level (left), with zooming into the carbon emission factors at machine level (top right and bottom right). Equations (27)–(32) were used to calculate the carbon emissions at machine level (top left and bottom right), with the heaviest contributing at machine level being the carbon emission factor for electrical energy (left), followed by carbon emission factor of lubricant consumption (top right), carbon emission factor of fixture consumption (bottom right), and carbon emission factor of tool consumption (bottom right)

Full size image
Fig. 11
figure 11

Tornado plot for deforming at system level (left), with zooming into the system level part (right). Overall, at system level, the heaviest contributing factor is the carbon emission factor of electrical energy, followed by carbon emission factor of steel production, carbon emission factor of Cu production, and carbon emission factor of lubricant production (right). The impact of carbon emissions of the machine tool level (e.g., lubricant consumption, tool consumption, and fixture consumption) is negligible

Full size image

5 Discussion

Carbon emissions in manufacturing are generated from consumption of electrical energy at process, machine (mainly by auxiliary equipment), and system level; consumption of tools and coolant liquids; production and disposal of material, tools, and liquids; and transportation of materials in general. From the analysis presented earlier (Sections 3 and 4), there are two main types of carbon emissions: (a) related to consumption of electrical energy and (b) related to production, disposal, or transportation of raw materials, tools, and fluids. From the sensitivity analysis in all three examples, the common source of carbon emissions is the electrical energy at process and machine levels. At system level, the situation is more complex and is highly dependent on the example. But even in these cases, the carbon emissions related to the electrical energy are one of the strongest contributing factors for the total carbon emissions at system level.

Most studies on carbon emission calculation are using a simplistic approach of reducing the energy consumption via optimization of the process parameters, without weighing the importance of other sources of carbon emissions or without reducing the carbon emission factor for electrical energy. In details, the majority of the papers list the energy requirements of the process and the auxiliary equipment for a specific process and then multiply the result with the carbon emission factor for the electrical energy. For the rest of the carbon emissions that are not directly involved with the consumption of electrical energy, equations related to the consumption of resources are used. In general, the focus is primarily on the reduction of the electrical energy consumption, followed by the optimization of the use of resources. Despite the popularity of this approach, it fails to either result in significant carbon footprint reduction or to identify which parts of the process will result in the highest reduction of carbon emissions. The latter is extremely important since the carbon emissions from the production of raw materials can be significantly larger than the carbon emissions from the electrical energy consumption, as it has been shown in the case studies above.

In addition, the current work has highlighted the difference of the impact of carbon emission factors on the final value of carbon footprint, based on whether the calculation is at process, machine tool, or system level. To the knowledge of the authors, this have never been addressed in literature, as most papers calculate carbon footprint at a predefined level, without comparing the difference of carbon emissions at process, machine tool, or system level. In this way, the studies fail to identify which carbon emission factor weighs the more at each manufacturing level, and they could potentially choose to reduce or optimize the component that bares the least significance on the overall carbon footprint.

Alternatively, a proper approach should first define the heaviest contributing steps and then should follow one the following ways to reduce carbon emission factors: (1) by reducing the carbon emission factors, (2) by improving the energy efficiency of the process, (3) by better planning the design of the product, and (4) by using less carbon-intensive tools and consumables.

This approach provides an easy-to-follow and generalized methodology for calculating carbon emissions in manufacturing. However, it requires ways of decreasing the carbon emission factors which may not be feasible at all times. For example, reduction of carbon emission factor for electrical energy can be achieved via use of alternative or renewable power sources which may not be always easy to implement. Reduction of carbon emission factor for the production of raw materials can be done via use of recycled sources, but this may affect the quality of the final product. Therefore, one must identify the steps that can be modified according to the specific industry and the quality of the final product.

As already described (Sections 3 and 4), the carbon emission factors are qualitative parameters because they are related to how energy/material/consumable/tool is produced, consumed, disposed, or transported, while the energy efficiency, product design, and machine/consumable specifications are quantitative parameters and are related to the application in question. The different categories described in the next subsections are related to (1) material needed for the manufacturing, (2) machine specification used in the specific application, and (3) electrical energy consumed. The discussion is built upon the sensitivity analysis described in the previous sections. Extensive data analysis is not part of this paper’s scope as this work provides a review of the literature and proposes a general framework to be used regardless the type of process. Future work on specific examples or processes will include further data analysis to identify the specific factors that contribute to carbon footprint, to define the optimizable parameters with the highest impact on carbon emissions and to provide process-specific alternatives of the general framework.

5.1 Material needed for manufacturing

Material needed for the manufacturing process is related to the raw or recycled material used as an input for each process. The effect of material selection is important, mainly because of the carbon emission factor for material production, but also because of quantitative parameters such as material removal rate in material removal process or mass of sand in primary forming process.

As seen from the tornado plots in the sensitivity analysis (Figs. 5 and 11), carbon emission factors of material production are among the heaviest contributing factors, with being the heaviest in material removal example and the second heaviest in the deforming example.

5.1.1 Quantitative

The quantitative parameters of the materials used in the calculation of carbon footprint are related to the design of the product, e.g., value of material removal and mass of chips in the material removal process (Mchip), energy required for pressing, mass of steel, and mass of Cu used in the deforming. In the primary forming sensitivity analysis, the power and rate of compression at the process levels are indirectly related to material, while the mass of sand used and the surface area of the final beam are the quantitative parameters at the machine level. At the system level, the fractions of sand production and sand reclamation are important for the carbon emission calculation.

5.1.2 Qualitative

The qualitative parameters related to the material are the carbon emission factors for material production and material disposal. From the sensitivity analysis, the factor for the carbon emission factor for material production has the highest impact on the value of the carbon emissions from the system, as shown in the sensitivity analysis of the material removal (0.9051) and second highest for the deforming process (0.1415 for Cu and 0.1665 for Steel). The qualitative parameters for the primary forming are the fraction of clay (ηcla) and sand (ηsan) which are strongly related to the final product specifications.

5.2 Machine specifications

Machine tool specifications are important for the carbon emissions at machine level, as shown in the tornado plots (Figs. 4 and 10). However, their contribution is not as strong as the contribution from electrical energy and material production at system level.

5.2.1 Quantitative

The quantitative parameters of the machine tool in the material removal process are mainly concerning the feed rate and the cutting rate of the machines, as well as the volume of coolant consumed and the mass of cutting tool, along with their lifetimes. In addition, the power consumed during the air cutting, basic running, and shutting down are also qualitative parameters related to the machine tool and affect the carbon emissions from the machine tool.

For the primary forming, the quantitative parameters are the power and rate of compression at process level, while power and rate at the rest of steps related to the machine tool level are included. At the system level, the power and rate of mixing, reclamation, and production are included. In addition, temperature and heat capacity of sand and water are linked to the carbon emissions at machine tool level from the machine/materials.

In cases of deforming, the quantitative parameters are the energy required for pressing, the time for forming, the lifetimes of forming tool and fixtures, and the volume of lubricant used.

5.2.2 Qualitative

The carbon emission factor of coolant production and disposal and of tool consumption are the qualitative factors of the machine tool that contribute to the carbon emissions. Carbon emission factor for coolant consumption is not a strong contributor the carbon emission, either at machine or system level, with sensitivity factors at 0.0058 and 0.0005, respectively. The carbon emission factor of coolant disposal is the lowest contributor to the system carbon emissions with sensitivity factor of 4*10−5. The carbon emission factor of tool consumption (0.3262) is the second largest sensitivity factor for the machine carbon emissions, while it is the third largest for the system carbon emissions (0.0286).

In the sensitivity example for primary forming of beams used in this review, the carbon emission factors are related to the use of coal (0.4849), crude oil (10−7), and diesel (0.1582). In the incremental sheet forming, the carbon emission factors for lubricant consumption, tool consumption, and usage of fixtures are the qualitative parameters. At both machine and system level, among the three carbon emission factors mentioned earlier, the strongest is the lubricant consumption, followed by usage of fixture and tool consumption.

5.3 Energy needed for manufacturing

In the non-phase changing processes (material removal and deforming), the tornado plots show that the carbon emission factor for electrical energy is one of the two top contributing factors (Figs. 3, 4, 5 for material removal and deforming (Figs. 9, 10, 11) at all levels (process, machine, system). In phase-changing processes (primary forming), there are also other sources for energy, mainly from coal, crude oil, nature gas, and diesel. As seen from the relevant tornado plots, at process (Fig. 6) and machine level (Fig. 7), electrical energy is still the main contributing factor, while the rest of the carbon emission factors are introduced at system level, with carbon emission factor of coal being the heaviest, surpassing even the contribution of carbon emission factor for electrical energy. Then the carbon emission factor of diesel follows, with crude oil and gas carbon emission factors contributing marginally to the carbon emissions of primary forming at system level (Fig. 8).

5.3.1 Quantitative

Regarding the material removal, there is an electrical energy factor for each process and material for the energy requirements (kc). In the primary forming example, the electrical energy consumed for clay and sand is related to the carbon emissions at system level, while in the deforming example, the energy consumed is used for annealing, bonding, and cutting at the machine level, besides the energy used for the process for both primary forming and deforming which is related to material and machine tool. There are also parameters indirectly related to electrical consumption, and these are related to material and machine specifications, but there is no straightforward relation between these parameters and the energy consumption. Most studies have been focusing in the making the processes more energy efficient ([18, 94]), which by its own is not enough to significantly decrease the carbon emission factors.

5.3.2 Qualitative

Carbon emission factor for electrical energy is calculated based on the contribution of coal, natural gas, and petroleum to the power grid, with the heaviest contribution coming from the coal. Therefore, a first step towards decreasing CEFelec is the reduction of coal as a power source while increasing the natural gas contribution.

Following Green Deal and EU carbon neutral 2050 directives, the power sources will become greener, therefore, decreasing carbon emission factor electrical energy. Notably, CEFelec(solar) = 6 gCO2e/KWh and CEFelec(wind) = 4 gCO2e/KWh.

Carbon emission factor of electrical energy is calculated from the power grid and the source of power (coal, natural gas and oil)

$$ {CEF}_{elec}=\eta \cdotp \left(112\cdotp \%C+49\cdotp \% NG+66\cdotp \%P\right) $$

The value of the carbon emission factor for electrical energy depends on the country’s emissions to produce electrical energy, which can differ from region to region. CEFelec can be decreased based on the way power is generated.

In the sensitivity analysis for the material removal, the electrical energy carbon emission factor is the only factor (sensitivity factor 1), while at machine level and system level, there are other factors; hence, the contribution of CEFelec is restricted, with sensitivity factors if 0.668 at machine level and 0.0585 for system level.

Regarding the primary forming example, similarly the carbon emission factor for electrical energy is the only factor for process and machine level, while its contribution drops at 0.357 at system level. There is also the gas consumption for certain parts of the primary forming, but it does not contribute to the carbon emissions at the system level (6*10−9).

For the deforming example, the carbon emission factor for electrical energy is the strongest factor (sensitivity factor 1), its influence decreasing at the following levels with sensitivity factors 0.981 for the machine level and 0.6445 at the system level.

6 Conclusions

Across all levels and processes examples, the energy-related carbon emissions are either the strongest or the second strongest contributing factors, therefore should be one of the first to be addressed on the way to carbon emission reduction. This can be approached by either using as less energy as possible, the usual approach in literature, or by considering the type of energy used and its environmental impact. The first approach is linked to a careful design of product and process planning and scheduling, along with using energy efficient equipment. The second approach is related to the carbon emission factors of energy consumption, which is how the energy is generated and what is their environmental impact. Using alternative power sources can decrease the carbon emission by up to 99%, therefore making a huge impact on the carbon emission factors, much more than simply improving the energy efficiency of processes and equipment.

Another important fraction of carbon emissions, if not more important than electrical energy, derives from the material selection. In terms of quantitative parameters, a better selection of raw material and a careful design of the process will affect the carbon emissions at system level. Regarding the qualitative parameters, the carbon emission factors for material production depend on whether the material is extracted from ore or it is recycled, with recycling materials having much lower carbon emission factors.

The remaining carbon emissions are related to auxiliary equipment and coolant/lubricant usage, production, and disposal. Their impact is not as strong as the first two categories, but they are related to machine specifications or characteristics. Therefore, reduction of these carbon emissions can be done via usage of alternative liquids (e.g., use of nanofluid in machining [95] or use of vegetable ester-based biodegradable oil [96]), aiming to decrease the relevant carbon emission factor or via careful design of process requiring the least amount of coolant. In addition, selection of appropriate auxiliary tools has their own impact on carbon emissions, with studies showing that material type and velocity of cutting tools affecting the carbon emissions of the overall process [97].

In an order of high to low importance/contribution to carbon emissions, the order is consumption of electrical energy, production of raw material, and consumption/disposal of tools and consumables. Therefore, if someone aims to decrease the carbon footprint of processes and products and create more sustainable processes and products, one must start by addressing and decreasing the electrical energy consumption and secondly by using recycled material, instead of extracting raw materials.