Prioritisation of drivers of sustainable additive manufacturing using best worst method

ABSTRACT

Globalisation and competition in the market have made industry practitioners to focus towards minimising resource consumption and even government is making strict regulations towards sustainable development. In this regard, usage of latest technologies such as Additive Manufacturing (AM) will have a positive benefit towards sustainability. AM processes are becoming sustainable from viewpoint of less material consumption, lesser operational cost, minimum handling and so on. So, it is important to analyse the potential drivers of sustainable AM process for its smooth adoption. In this regard, the purpose of the article is to present a study on prioritising drivers of Sustainable Additive Manufacturing (SAM) using Best Worst Method (BWM). Forty drivers are being analysed from eight perspectives. Data have been collected from experts and BWM has been used for analysing the drivers. The drivers are analysed in respective categories and then global score for each driver has been computed to find the overall weightage and rating of each driver. Key drivers are identified as eco design, green innovation and energy conservation. The practical and managerial implications are being highlighted.

KEYWORDS:

• Additive manufacturing
• sustainable manufacturing
• drivers
• best worst method
• multi criteria decision making

1. Introduction

Additive manufacturing (AM) is the process of fabricating a part or a component by adding material in a layer-by-layer manner. AM is also called as layered manufacturing. AM has the ability to fabricate complex structures with lesser time and good material properties (Rejeski, Zhao, and Huang 2018). In the recent days, AM processes are expected to be sustainable (Chen et al. 2015; Laverne et al. (2019). Sustainable AM (SAM) has advantages from economy point of view, consumes less material which leads to reduction in operational cost (Ullah et al. 2013; Weller, Kleer, and Piller 2015). It also reduces energy consumption and emissions resulting in sustainable benefits (Yang and Li 2018). AM reduces greenhouse emissions by 27–58% (Kreiger and Pearce 2013). It brings customised parts which are less likely to dispose by consumers. It brings freedom in design, so that customised complex parts can be made easily within shortest possible time (Ford and Despeisse 2016). There are several business benefits in implementing SAM such as, it increases sales by meeting social and environmental expectations better than competitors, it reduces waste thereby increasing productivity and it enhances reputation of the firm, employee morale and better community relations. It consumes approximately 90% lesser material as compared to conventional manufacturing (Mehrpouya et al. 2019). Assessing sustainability of technologies is very essential nowadays to identify and select best sustainable technology among all available technologies which have potential to reduce environmental impact and is economically feasible and provide benefits to society (Jiang et al. 2019). In order to facilitate effective implementation of SAM, drivers need to be analysed. In this regard, this study aimed at analysing the drivers of SAM using multi criteria decision-making (MCDM) based best worst method (BWM) method. It has several advantages over other MCDM methods like, small group pairwise comparison, and more consistent (Ahmadi, Kusi-Sarpong, and Rezaei 2017).

The research questions addressed in the present study are:

• What are the various drivers influencing SAM?

• How to formulate the approach of analysing SAM drivers as multiple criteria decision-making problem?

• How to arrive at the priority order of SAM drivers using Best Worst Method (BWM)?

In this viewpoint, this article analyses the drivers of SAM from eight perspectives namely, Materials, Management and stake holders, Customer/ Supplier/ Competitor¸ Collaboration and trends, Technology, Standards and regulations, Design and Process. BWM is used in the present analysis. Based on the study, key drivers are found and the practical inferences are presented.

2. Literature review

SCOPUS database was chosen to collect relevant articles in the field of AM with sustainable domains. Several articles have been collected from SCOPUS database and important articles have been shortlisted based on abstract reading. The shortlisted articles have been reviewed thoroughly. The review is presented in the subsequent two subsections:

2.1 Review on studies of drivers of green/sustainable manufacturing

Bey, Hauschild, and McAloone (2013) did a systematic detailed international survey with reference to product development and manufacturing firms. The study recognised vital drivers for deploying product life cycle-based environmental strategies and also indicated the barriers and obstacles to be dealt with. With reference to the study, various suggestions for manufacturing firms as well for policy makers to deal with effective deployment of strategic environmental goals in manufacturing firms. The vital barriers include lack of information on environmental impact, lack of expert knowledge, and lack of allocated resources. Also, the vital drivers both for inducement and sustenance deployment include legislative demands, customer demands and expected competitive advantage. Garg, Luthra, and Haleem (2014) made an attempt to recognise and assess drivers in deploying Sustainable Manufacturing (SM) in Indian scenario. Nine drivers for effective deployment were identified from literature review. Decision Making Trial and Evaluation Laboratory (DEMATEL) was applied to assess and group the drivers into cause and effect categories. Among the drivers, five belong to cause group and four belong to effect group. From the findings, ‘societal pressure & public concern’ was found the most critical driver, whereas ‘corporate image & benefits’ was found least critical driver. It was mentioned that the developed model would help the practitioners in setting priorities of these drivers towards effective deployment of SM in Indian firms. Also, it was indicated that the implications would enable researchers to effectively understand issues pertaining to SM. Mittal and Sangwan (2014) mentioned that green manufacturing implementation in industries would facilitate overcoming environmental issues like climate change and global warming. They mentioned that green manufacturing deployment was bounded with many challenges. The study ends with ranking the motivating factors using fuzzy TOPSIS based on environmental, economical and social perspectives. Green manufacturing implementation could be done with collaborative effort of government and industries in a strategic manner. There exists a need to recognise the scope of various drivers enabling green manufacturing implementation in industries. The examination of green manufacturing drivers and their prioritisation was done based on triple bottom line dimensions.

Nordin, Ashari, and Hassan (2014) aimed to examine the drivers and barriers of SM deployment in Malaysian scenario. Data collection was done through questionnaire being self-administered. They found that increment in overall implementation cost as vital barrier for deploying SM practices; whereas environmental regulation and top management commitment were the key drivers. Based on the study, the prioritised barriers and drivers could help industries for enhancing manufacturing practices with minimal negative impact to environment. Also, it was mentioned that key focus to be assigned to motivating and hindering factors to improve successful deployment. Ghazilla et al. (2015) mentioned that the drivers and barriers of deploying green manufacturing practices in SMEs were different from that of large enterprises as SMEs lack data, resources, experience to implement green initiatives. It was mentioned that as SMEs contribute a significant role in nation’s economic growth, it is vital to recognise the drivers and barriers which motivate and hinder the deployment of green manufacturing practices in SMEs. The study reported basic findings of the drivers and barriers relevant to SMEs in deploying green practices in Malaysian context. Delphi-based survey approach was applied to explore, recognise and validate the drivers and barriers of green manufacturing from expert opinion. It was indicated that the study would expedite the deployment of green manufacturing practices in Malaysian SMEs. It was mentioned that during further stages, detailed views from relevant stakeholders would be obtained from green manufacturing practices deployment. Govindan, Diabat, and Shankar (2015) examined the need for recognising 12 mutual drivers of green manufacturing through literature review, industrial managers and expert opinion views. They administered the questionnaire among 120 leading firms in southern India and obtained pairwise comparison among those drivers. They further analysed the pairwise comparison data using fuzzy Multi Criteria Decision Making (MCDM) method. The results generated were further analysed using two stage approach: applying various defuzzification methods and assigning different weights to the top prioritised drivers. From the results, compliance with regulations was found to be the most essential driver, stakeholders and customers form the second and third essential drivers of green manufacturing. The results were in coincidence with literature observation. It was also mentioned that the study could analyse the barriers that hinder the adoption of green practices in SMEs as well in house hold firms.

Shankar, Kumar, and Kannan (2016) analysed the drivers of advanced SM through a framework that was being proposed and further validated using a case study. The goal was to investigate the need for inclusion of sustainable strategies into advanced manufacturing system. They collected widely referred drivers from literature, validated through expert opinion, and further analysed using MCDM tool, AHP. Based on the study, quality ranked first followed by financial benefit; market capabilities were found to be the least driver of advanced SM. Manufacturing professions can synthesise the top rank driver and further adopt to deploy advanced SM. Aboelmaged (2018) examined the impact of technological, organisational and environmental drivers on SM practices and further the influence of them on competitive capabilities. The model was validated using partial least square structural equation modelling approach in line with data gathered from Egyptian SMEs. The key results imply that environmental pressure from stake holders, management support and employees engagement influence positively SM practices. The research was divided into two major streams: SM research stream pertaining to sustainable and green manufacturing and the second stream was connected with vital sustainability issues indirectly connected with SM. The model was empirically validated based on data from 238 owners and managers belonging to various Egyptian SMEs. It was mentioned that from a managerial viewpoint, the research provides policy makers and managers a reference framework to enhance SM practices in the viewpoint of SMEs. Moktadir et al. (2018) attempted to deploy SM practices as a segment of green supply chain initiatives about leather industries in Bangladesh. The main focus of the study was to evaluate, prioritise and rank the drivers of SM practices. Grey theory and matrix approach were applied to analyse the drivers. The results depicted that knowledge of circular economy was vital to deploy SM practices in leather industries in Bangladesh. The study assisted managers to appropriately formulate strategies for optimum resource utilisation as well for waste reduction in the perspective of circular economy. Sustainable manufacturing was integrated with industry 4.0 technologies such as AM to enhance the performance of manufacturing firm (Machado, Winroth, and Ribeiro da Silva 2020). Sustainable manufacturing strategies were recognised and prioritised by Ocampo et al. (2020) for smooth adoption of SM practices in manufacturing sector.

Based on the literature it is found that many researches have been conducted on drivers analysis in terms of sustainable manufacturing. Various drivers identified are legislative demands, customer demands, expected competitive advantage (Bey, Hauschild, and McAloone 2013), societal pressure & public concern (Garg, Luthra, and Haleem 2014), environmental regulation and top management commitment (Nordin, Ashari, and Hassan 2014), and compliance with regulations (Govindan, Diabat, and Shankar 2015).

2.2 Review on studies of drivers of Additive Manufacturing/ sustainable AM

Agrawal and Vinodh (2019) developed a structural model in line with TISM approach to analyse the factors influencing SAM. They considered 20 factors based on literature analysis, gathered expert opinion and further developed the structural model. They indicated various benefits of AM from sustainability perspective namely: lesser material consumption, scope for redesign, generation of light weight structures. With reference to the developed model, the dominant factors were identified which could help the practitioners for evolving long-term plan for improving sustainability of AM technologies, further MICMAC approach was done and ten factors were found to be dependent and ten factors belong to driving group respectively. Niaki, Torabi, and Nonino (2019) mentioned the reasons for adoption of AM in various industrial sector which depicted potential advantages in terms of sustainability orientation. The study aimed to recognise and prioritise the determinants of its implementation as well to clarify the need of sustainability benefits. The study deployed a two round survey to enable empirical results into present viewpoint of AM deployment from practitioners viewpoint. Based on the analysis of first round survey, it indicated economic benefits were the primary reason whereas societal and environmental benefits were least drivers. An analysis of second round survey indicated the capability of AM for manufacture of complex parts, enabling customisation and inculcating creativity and innovation were the vital drivers for AM adoption.

From the literature study, it can be found that very few studies are available pertaining to additive manufacturing analysis. Hence, there exist a need to analyse drivers of SAM to strengthen the theories of SAM.

2.3 Research gaps

With reference to literature analysis, it has been found that there are reasonable number of studies on analysis of SM drivers and limited studies with regard to AM drivers. In line with the technological changes to move towards SAM with the focus on developing products with minimal environmental impact through the generation of light weight structure, lattice geometries, near net shape parts manufacturing and end used parts fabrication directly, SAM is found to be more significant. In order to facilitate the deployment of SAM, the analysis of drivers is vital. In this context, the present study illustrates an attempt to analyse the drivers of SAM using MCDM approach. In order to facilitate small group pairwise comparison and to derive consistent result, BWM is used as solution methodology for analysis of SAM drivers.

3. Methodology

The goal of this study is to prioritise the drivers of SAM. MCDM is an appropriate method to analyse the drivers and identify the prominent drivers (Ahmadi, Kusi-Sarpong, and Rezaei 2017). In this study, Best Worst Method (BWM) is used for prioritisation of SAM drivers. The methodological flowchart for prioritising SAM drivers is presented in Figure 1.

Figure 1. Methodological steps

3.1 Best Worst Method (BWM)

BWM has been used in this study to analyse the drivers of SAM. The reason for selecting the methodology is that, it has several advantages over other MCDM methods like, small group pairwise comparison is analysed in this method whereas full pairwise comparison matrix in other MCDM method. Secondly, the result associated with it is more consistent compared to other MCDM approaches (Ahmadi, Kusi-Sarpong, and Rezaei 2017). Third it involves comparison of best criteria with other criteria and comparison of other criteria with worst criteria, thereby elimination of secondary comparison matrix which are required in other MCDM methods (Shojaei, Haeri, and Mohammadi 2018). It is already being used in various areas for analysis. Table 1 shows the previous studies on BWM.

Table 1. Previous studies on BWM application in Sustainability domain

Research Study	Focus
Rezaei et al. (2016)	Evaluating best supplier by considering sustainability criteria
Ahmadi, Kusi-Sarpong, and Rezaei (2017)	Evaluating social aspect of sustainability in supply chain domain.
Ahmad et al. (2017)	Analysed the external forces influencing the sustainability of oil and gas supply chain
Zhao, Guo, and Zhao (2018)	Analysed and ranked the eco industrial parks based on its benefits considering sustainability and circular economy in mind.
Malek and Desai (2019)	Analysed the barriers of SM for an Indian manufacturing firm.

The steps of BWM (Rezaei et al. 2016) are presented below:

Step 1: First step is to identify criteria for decision making from literature review or from survey.

Step 2: Second step is to analyse the criteria and recognise the best criteria and worst criteria. The best criteria signify the most vital and desirable criteria whereas the worst criteria signify the least important and least desirable criteria. This identification of best and worst criteria can be done with the help of experts or decision makers

Step 3: Third step is to benchmark the best criteria with other criteria. A rating of 1 to 9 need to be given for comparison where 1 signifies equal importance and 9 signifies extreme importance with respect to best criteria. This is called as comparison of Best–to-Others (BO)

It is represented as vector $A_B$ = ($a_{B1},a_{B2},\dots \dots \dots .a_{Bn}$)

Where $a_{Bj}$ indicates the comparison of identified best criteria B over other criteria j.

Step 4: Fourth step is to benchmark other criteria with worst criteria. A rating of 1 to 9 need to be given for comparison where 1 signifies equal importance and 9 signifies extreme importance with respect to worst criteria. This is called as comparison of Others–to-Worst (OW)

It is represented as vector $A_W$ = ($a_{1W},a_{2W},\dots \dots \dots \dots a_{nW}$)^T

Where $a_{jW}$ implies the comparison of other criteria j over identified worst criteria W.

Step 5: In this step, optimal weight of each criterion (W₁*, W₂* … … W_n*) need to be evaluated.

For each pair ratio of W_B/W_j and W_j/Ww, the optimum weight of criteria should satisfy the below mentioned condition:

$\frac{W_B}{W_j}$ =$a_{Bj}$ and $\frac{W_j}{W_W}$ =$a_{jW}$

Therefore, to identify the optimum weight of criteria, maximum absolute difference of $\left|\frac{W_B}{W_j}-a_{Bj}\right|$ and $\left|\frac{W_j}{W_W}-a_{jW}\right|$ for all criteria j to be minimised. The problem can be represented in mathematical form as:

(1) $Minma{x}_j\left\{\left|\frac{W_B}{W_j}-a_{Bj}\right|,\left|\frac{W_j}{W_W}-a_{jW}\right|\right\}$(1)

Subjected to:

$\sum _{j=1}^n{W}_j=1$

$W_j\ge 0forallj$

Hence equation 1(1) $Minma{x}_j\left\{\left|\frac{W_B}{W_j}-a_{Bj}\right|,\left|\frac{W_j}{W_W}-a_{jW}\right|\right\}$(1) can be rewritten as:

(2) $\mathrm{M}\mathrm{i}\mathrm{n}\mathrm{m}\mathrm{a}{\mathrm{x}}_{\mathrm{j}}\xi$(2)

Subjected to:

$\left|\frac{W_B}{W_j}-a_{Bj}\right|\le \xi forallj$

$\left|\frac{W_j}{W_W}-a_{jW}\right|\le \xi forallj$

$\sum _{j=1}^n{W}_j=1$

$W_j\ge 0forallj$

By solving equation 2(2) $\mathrm{M}\mathrm{i}\mathrm{n}\mathrm{m}\mathrm{a}{\mathrm{x}}_{\mathrm{j}}\xi$(2) , the optimum weights of each criteria (W₁*, W₂* … … W_n*) are obtained and ξ. ξ is used for evaluating consistency ratio. Lower value of ξ signifies higher consistency level.

The consistency ratio of the BWM can be evaluated using equation 3(3) $ConsistencyRatio=\frac{\xi }{ConsistencyIndex}$(3) .

(3) $ConsistencyRatio=\frac{\xi }{ConsistencyIndex}$(3)

Value of consistency index is taken from Table 2 (Rezaei 2015).

Table 2. Consistency index (CI)

$a_{BW}$	1	2	3	4	5	6	7	8	9
CI value	0	0.4	1	1.63	2.3	3	3.75	4.47	5.23

4. Case study

The goal of the present study is to identify the prominent drivers of SAM and to prioritise those drivers. BWM has been used in the present study to analyse the drivers. From SCOPUS database, articles were collected to identify drivers of SAM. Further, the identified drivers were consulted with experts. Based on the consultation with experts, 40 prominent drivers of SAM have been shortlisted. The shortlisted drivers have been grouped into eight different categories namely: Materials, Management and stake holders, Customer/ Supplier/ Competitor¸ Collaboration and trends, Technology, Standards and regulations, Design and Process. The identified drivers are depicted in Table 3.

Table 3. Drivers of sustainable additive manufacturing

Category	Driver	References
Material (M)	Optimum resources utilisation (M1)	Govindan, Diabat, and Shankar (2015); Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016)
	Reusing and recycling materials and packaging (M2)	Moktadir et al. (2018); Nordin, Ashari, and Hassan (2014)
	Less material wastage (M3)	Mani, Lyons, and Gupta (2014); Rahimifard et al. (2009)
Management and Stake Holders (M & SH)	Proper training and education (MSH1)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016); Vinodh, Ramesh, and Arun (2016)
	Incentives (MSH2)	Mittal and Sangwan (2014)
	Stake holder participation (MSH3)	Vinodh, Ramesh, and Arun (2016)
	Management support (MSH4)	Vinodh, Ramesh, and Arun (2016)
	Employee commitment towards sustainability (MSH5)	Vinodh, Ramesh, and Arun (2016)
Customer/ Supplier/ Competitor (CSC)	Customers’ expectations (CSC1)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016)
	Supplier commitment (CSC2)	Ageron, Gunasekaran, and Spalanzani (2012)
	Company image (CSC3)	Govindan, Diabat, and Shankar (2015); Searcy et al. (2012)
	Public pressure (CSC4)	Mittal and Sangwan (2014)
	Supply chain pressure (CSC5)	Mittal and Sangwan (2014)
	Competitor pressure towards greening (CSC6)	Moktadir et al. (2018); Ghazilla et al. (2015)
Collaboration and Trends (C & T)	Market trends (CT1)	Govindan, Diabat, and Shankar (2015); Searcy et al. (2012)
	Economic benefits (CT2)	Ghazilla et al. (2015); Nordin, Ashari, and Hassan (2014); Moktadir et al. (2018); Vinodh (2011)
	Collaboration between organisations (CT3)	Nordin, Ashari, and Hassan (2014); Moktadir et al. (2018)
	Long-term survival in market (CT4)	Nordin, Ashari, and Hassan (2014)
	Collaborative decision making (CT5)	Govindan, Diabat, and Shankar (2015); Vinodh, Ramesh, and Arun (2016)
Technology (T)	Green innovation (T1)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016);
	Energy conservation (T2)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016)
	Technology competence (T3)	Vinodh, Ramesh, and Arun (2016)
	Technology adaptability (T4)	Niaki, Torabi, and Nonino (2019)
	Time to develop new product (T5)	Raymond (2005)
	Effective visual control (T6)	Vinodh, Ramesh, and Arun (2016)
Standards and Regulations (S & R)	Environmental conservation (SR1)	Govindan, Diabat, and Shankar (2015); Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016)
	Emissions and global climate (SR2)	Moktadir et al. (2018); Siemieniuch, Sinclair, and Henshaw (2015)
	Safety and security (SR3)	Moktadir et al. (2018); Siemieniuch, Sinclair, and Henshaw (2015)
	Operational safety (SR4)	Wang et al. (2015);
	Compliance with regulations (SR5)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016);
Design (D)	Light weight parts (D1)	Khorram Niaki and Nonino (2017)
	Eco design (D2)	Govindan, Diabat, and Shankar (2015); Laverne et al. (2019)
	Flexibility in producing geometry (D3)	Mani, Lyons, and Gupta (2014)
	Ease of producing complex structure (D4)	Mani, Lyons, and Gupta (2014)
Process (P)	Cost saving (P1)	Niaki, Torabi, and Nonino (2019);
	Time saving (P2)	Niaki, Torabi, and Nonino (2019)
	On demand manufacturing (P3)	Attaran (2017)
	Enables mass customisation (P4)	Attaran (2017)
	No specialised tooling required (P5)	Mani, Lyons, and Gupta (2014)
	Quality (P6)	Bey, Hauschild, and McAloone (2013); Shankar, Kumar, and Kannan (2016)

After identifying the drivers of SAM, next step is to collect data from experts. Four experts have been selected for collecting data. The experts are having high competence in the field of sustainability and AM with more than 10 years of field experience.

Their responses have been collected through email. First the experts were requested to recognise the best driver and worst driver in all categories. Then experts were requested to benchmark best driver to all other drivers to generate vector BO. Similarly, experts were asked to compare other drivers with worst driver to generate vector OW.

All input data collected are presented in Appendix section.

Sample data collection for one category of drivers is presented in Tables 4, 5 and Table 6.

Table 4. Best and worst driver specified by expert 1–4 for material category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Optimum resources utilisation	1,4	2
Reusing and recycling materials and packaging	2	3
Less material wastage	3	1,4

Table 5. Comparison of Best driver with other drivers (BO) by expert 1–4 for material category

Expert	Best to Others	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage
E1	Optimum resources utilisation	1	3	7
E2	Reusing and recycling materials and packaging	7	1	3
E3	Less material wastage	5	7	1
E4	Optimum resources utilisation	1	4	9

Table 6. Comparison of other driver with worst drivers (OW) by expert 1–4 for material category

Expert	E1	E2	E3	E4
Others to Worst	Less material wastage	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage
Optimum resources utilisation	7	1	3	9
Reusing and recycling materials and packaging	5	7	1	5
Less material wastage	1	5	7	1 R

After collecting input data from experts, the weights of each driver have been computed using equations 1(1) $Minma{x}_j\left\{\left|\frac{W_B}{W_j}-a_{Bj}\right|,\left|\frac{W_j}{W_W}-a_{jW}\right|\right\}$(1) and 2(2) $\mathrm{M}\mathrm{i}\mathrm{n}\mathrm{m}\mathrm{a}{\mathrm{x}}_{\mathrm{j}}\xi$(2) as presented in methodology section. To evaluate the weights of each drivers, average of all expert weights has been taken. Table 7 shows the weights of sample drivers of material category calculated using BWM. The consistency ratio also has been calculated using equation 3(3) $ConsistencyRatio=\frac{\xi }{ConsistencyIndex}$(3) (Rezaei et al. 2016).

Table 7. Weights of driver along with consistency for material category

Expert	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage	ξ^L
E1	0.6615	0.2615	0.0769	0.1231
E2	0.0769	0.6615	0.2615	0.1231
E3	0.1688	0.0909	0.7403	0.1039
E4	0.7222	0.2111	0.0667	0.1222
Final Weight	0.4074	0.3063	0.2864	0.1181

Consistency Ratio (CR) = $\frac{\xi }{ConsistencyIndex\left(CI\right)}$

CR = $\frac{0.1181}{1}$ and CI (n =3) = 1

CR = 0.1181

Table 8(b). Weights of drivers along with consistency for material category

Expert	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage	ξ^L
E1	0.6615	0.2615	0.0769	0.1231
E2	0.0769	0.6615	0.2615	0.1231
E3	0.1688	0.0909	0.7403	0.1039
E4	0.7222	0.2111	0.0667	0.1222
Final weight	0.4074	0.3063	0.2864	0.1181
Rank	1	2	3

CI (n =3) = 1,

CR = 0.1181/1

CR = 0.1181

Table 8(c). Weights of drivers along with consistency for management and stake holders category

Expert	Proper training and education	Incentives	Stake holder participation	Management support	Employee commitment towards sustainability	ξ^L
E1	0.5084	0.0446	0.0879	0.1539	0.2052	0.1070
E2	0.1539	0.0879	0.0446	0.2052	0.5084	0.1070
E3	0.0424	0.0836	0.4836	0.1952	0.1952	0.1018
E4	0.1952	0.0424	0.0836	0.4836	0.1952	0.1018
Final weight	0.2250	0.0646	0.1749	0.2595	0.2760	0.1044
Rank	3	5	4	2	1

CI (n =5) = 2.3,

CR = 0.1044/2.3

CR = 0.045

Table 8(d). Weights of drivers along with consistency for customer/ supplier/ competitor category

Expert	Customers’ expectations	Supplier commitment	Company image	Public pressure	Supply chain pressure	Competitor pressure towards greening	ξ^L
E1	0.4406	0.1334	0.0762	0.0387	0.1334	0.1778	0.0928
E2	0.1702	0.0370	0.4219	0.0730	0.1277	0.1702	0.0888
E3	0.1702	0.1277	0.1702	0.0730	0.0370	0.4219	0.0888
E4	0.1494	0.1494	0.1494	0.3701	0.0325	0.1494	0.0779
Final weight	0.2326	0.1119	0.2044	0.1387	0.0827	0.2298	0.0871
Rank	1	5	3	4	6	2

CI (n =6) = 3,

CR = 0.0871/3

CR = 0.029

Table 8(e). Weights of drivers along with consistency for collaboration and trends category

Expert	Market trends	Economic benefit	Collaboration between organisations	Long term survival in market	Collaborative decision making	ξ^L
E1	0.1377	0.1836	0.4551	0.1836	0.0399	0.0958
E2	0.0418	0.4770	0.1444	0.1444	0.1925	0.1004
E3	0.0399	0.1836	0.1836	0.1377	0.4551	0.0958
E4	0.1836	0.4551	0.1836	0.0399	0.1377	0.0958
Final weight	0.1008	0.3248	0.2417	0.1264	0.2063	0.0970
Rank	5	1	2	4	3

CI (n =5) = 2.3,

CR = 0.0970/2.3

CR = 0.042

Table 8(f). Weights of drivers along with consistency for technology category

Expert	Green innovation	Energy conservation	Technology competence	Technology adaptability	Time to develop new product	Effective visual control	ξ^L
E1	0.4219	0.1702	0.1277	0.1702	0.0730	0.0370	0.0888
E2	0.1702	0.1702	0.4219	0.1277	0.0370	0.0730	0.0888
E3	0.1702	0.4219	0.1702	0.1277	0.0730	0.0370	0.0888
E4	0.4673	0.1886	0.0808	0.0410	0.1414	0.0808	0.0984
Final weight	0.3074	0.2377	0.2002	0.1167	0.0811	0.0570	0.0912
Rank	1	2	3	4	5	6

CI (n =6) = 3,

CR = 0.0912/3

CR = 0.0304

Table 8(g). Weights of drivers along with consistency for standards and regulations category

Expert	Environmental conservation	Emissions & global climate	Safety & security	Operational safety	Compliance with regulations	ξ^L
E1	0.5084	0.2052	0.0446	0.0879	0.1539	0.1070
E2	0.1952	0.1952	0.0424	0.0836	0.4836	0.1018
E3	0.1844	0.4664	0.1106	0.0542	0.1844	0.0868
E4	0.5084	0.1539	0.0879	0.0446	0.2052	0.1070
Final weight	0.3491	0.2552	0.0714	0.0676	0.2568	0.1007
Rank	1	3	4	5	2

CI (n =5) = 2.3,

CR = 0.1007/2.3

CR = 0.0437

Table 8(h). Weights of drivers along with consistency for design category

Expert	Light weight parts	Eco design	Flexibility in producing geometry	Ease of producing complex structure	ξ^L
E1	0.2425	0.6009	0.1039	0.0527	0.1265
E2	0.6009	0.2425	0.0527	0.1039	0.1265
E3	0.0527	0.2425	0.6009	0.1039	0.1265
E4	0.2425	0.6009	0.0527	0.1039	0.1265
Final weight	0.2847	0.4217	0.2026	0.0911	0.1265
Rank	2	1	3	4

CI (n =4) = 1.63

CR = 0.1265/1.63

CR = 0.0776

Table 8(i). Weights of drivers along with consistency for process category

Expert	Cost saving	Time saving	On demand manufacturing	Enables mass customisation	No specialised tooling required	Quality	ξ^L
E1	0.1211	0.1614	0.1614	0.4000	0.1211	0.0351	0.0842
E2	0.1261	0.1261	0.4168	0.1261	0.1682	0.0366	0.0878
E3	0.0370	0.1277	0.1702	0.1702	0.4219	0.0730	0.0888
E4	0.0370	0.4219	0.1702	0.1702	0.1277	0.0730	0.0888
Final weight	0.0803	0.2093	0.2297	0.2166	0.2097	0.0544	0.0874
Rank	5	4	1	2	3	6

CI (n =6) = 3,

CR = 0.0874/3

CR = 0.0291

5. Results and discussions

In the present study, 40 drivers of SAM have been considered from literature review. A MCDM based BWM approach has been systematically applied to analyse the drivers of SAM. First, the drivers have been grouped in eight categories and weights of drivers in individual categories are calculated. Then weights of categories are also calculated by applying the same procedure. Then finally the individual weight of drivers is multiplied with their respective category weight to get the global weight of each driver. The weights and ranking of driver in each category are shown in Tables 8 (a) −8(i).

Table 8(a). Weights of Categories along with consistency

Expert	Material	Management and stake holders	Customer/ supplier/ competitor	Collaboration and trends	Technology	Standards and regulations	Design	process	ξ^L
E1	0.1369	0.0298	0.0587	0.1027	0.1369	0.0587	0.3394	0.1369	0.07144
E2	0.1312	0.0562	0.0984	0.0285	0.3251	0.0984	0.1312	0.1312	0.06843
E3	0.1312	0.0562	0.0285	0.0984	0.3251	0.0984	0.1312	0.1312	0.06843
E4	0.1369	0.0587	0.0587	0.1027	0.1369	0.0298	0.1369	0.3394	0.07144
Final weight	0.1331	0.0502	0.0611	0.0831	0.2310	0.0713	0.1847	0.1847	0.0699

CI (n =8) = 4.47,

CR = 0.0699/4.47

CR = 0.015

The global weight of each driver and the priority ranking are presented in Table 9.

Table 9. Prioritised ranking of drivers of Sustainable Additive Manufacturing

Category	Category weight	Driver code	Weight	Global weight	Rank
Material (M)	0.1331	M1	0.4074	0.0542	4
		M2	0.3063	0.0408	8
		M3	0.2864	0.0381	12
Management and Stake Holders (M & SH)	0.0502	MSH1	0.225	0.0113	30
		MSH2	0.0646	0.0032	40
		MSH3	0.1749	0.0088	33
		MSH4	0.2595	0.0130	28
		MSH5	0.276	0.0139	26
Customer/ Supplier/ Competitor (CSC)	0.0611	CSC1	0.2326	0.0142	24
		CSC2	0.1119	0.0068	36
		CSC3	0.2044	0.0125	29
		CSC4	0.1387	0.0085	34
		CSC5	0.0827	0.0050	38
		CSC6	0.2298	0.0140	25
Collaboration and Trends (C & T)	0.0831	CT1	0.1008	0.0084	35
		CT2	0.3248	0.0270	14
		CT3	0.2417	0.0201	17
		CT4	0.1264	0.0105	31
		CT5	0.2063	0.0171	21
Technology (T)	0.2310	T1	0.3074	0.0710	2
		T2	0.2377	0.0549	3
		T3	0.2002	0.0462	6
		T4	0.1167	0.0269	15
		T5	0.0811	0.0187	18
		T6	0.057	0.0132	27
Standards and Regulations (S & R)	0.0713	SR1	0.3491	0.0249	16
		SR2	0.2552	0.0182	20
		SR3	0.0714	0.0051	37
		SR4	0.0676	0.0048	39
		SR5	0.2568	0.0183	19
Design (D)	0.1847	D1	0.2847	0.0526	5
		D2	0.4217	0.0779	1
		D3	0.2026	0.0374	13
		D4	0.0911	0.0168	22
Process (P)	0.1847	P1	0.0803	0.0148	23
		P2	0.2093	0.0386	11
		P3	0.2297	0.0424	7
		P4	0.2166	0.0400	9
		P5	0.2097	0.0387	10
		P6	0.0544	0.0100	32

From the analysis, it is found that ‘Technology’ is the most important category among all eight categories. Followed by ‘Design’, ‘Process’, ‘Material’, ‘Collaboration and Trends’, ‘Standards and Regulations’, ‘Customer/ Supplier/ Competitor’, and ‘Management and Stake Holders’. Manufacturing sector generates heavy environmental waste and consumes more resources which is a point of concern for industries, researchers and everyone. Technology is the most important factor associated with waste generation and resource consumption. Proper focus needs to be given to technology domain to reduce environmental impact and making process more sustainable.

In technology category, ‘Green innovation’ driver ranked first followed by ‘Energy conservation’, ‘Technology competence’, ‘Technology adaptability’, ‘Time to develop new product’, and ‘Effective visual control’. In earlier days, innovation happens by considering only technology, but nowadays industries are more prone towards sustainable development and innovation also called as green innovation.

In Design category, ‘Eco design’ ranked first followed by ‘Light weight parts’, ‘Flexibility in producing geometry’ and ‘Ease of producing complex structure’. Industries and practitioners are focusing more on design aspects of product development so as to make eco design parts by reducing unnecessary wastage, and material consumption. With the help of AM technology, complex structure can be made easily.

In Process category, ‘On demand manufacturing’ ranked first followed by ‘Enables mass customisation’, ‘No specialised tooling required’, ‘Time saving’, ‘Cost saving’, and ‘Quality’. AM technology has ability to fabricate part on demand. With AM technology, it became easy to provide replacement parts at remote locations also in less time.

Under Material category, ‘Optimum resources utilisation’ driver ranked first followed by ‘Reusing and recycling materials and packaging’, and ‘Less material wastage’. Optimum utilisation of resources is the key factor of AM technologies. AM has advantage of fabricating the parts in different orientations so as to minimise material consumption.

In Collaboration and Trends category, ‘Economic benefit’ driver ranked first followed by ‘Collaboration between organisations’, ‘Collaborative decision making’, ‘Long-term survival in market’, and ‘Market trends’. The implementation of sustainable practices along with AM technologies in manufacturing sector will lead to enhance system capabilities followed by providing economic benefits with minimal resource consumption.

Under Standards and Regulations, ‘Environmental conservation’ driver ranked first followed by ‘Compliance with regulations’, ‘Emissions & global climate’, ‘Safety & security’, and ‘Operational safety’. Day by day depletion of natural resources is a major issue and is very much important for environmental conservation. It is the main driving force for implementing SAM practices.

Under Customer/ Supplier/ Competitor category, ‘Customers’ expectations’ driver ranked first followed by ‘Competitor pressure towards greening’, ‘Company image’, ‘Public pressure’, ‘Supplier commitment’, and ‘Supply chain pressure’. Nowadays, customers are demanding ecofriendly and sustainable products. They are more concerned about environmental issues, which needs manufacturing firm to implement SM practices in AM technologies.

Under Management and Stake Holders category, ‘Employee commitment towards sustainability’ ranked first followed by ‘Management support’, ‘Proper training and education’, ‘Stake holder participation’, and ‘Incentives’. Full-fledge involvement of employee is very much important in successful implementation of SAM processes.

From Table 9, it can be seen that Eco design, Green innovation, Energy conservation, Optimum resource utilisation, and Light weight parts are the top five global drivers among all collected 40 drivers with global weight of 0.0779, 0.0710, 0.0549, 0.0542 and 0.0526 respectively. Eco-design concept is the most important driver as it leads to optimised material and energy consumption (Peng et al. 2018).

5.1 Sensitivity analysis

Sensitivity analysis is done for checking the biasness of the result and to check the impact of highest weightage driver on other drivers. Sensitivity analysis is done by variation of weights of all driver criteria in amount to variation in highest weightage driver criteria. Appendix Table A10 shows the variation of driver criteria weights corresponds to change in highest weightage driver criteria i.e, Design with weight 0.1847. By varying the weight of design criteria from 0.1 to 0.9, other driver criteria weights are identified as represented in Appendix Table A10. Appendix Table A11 shows the ranking of drivers based on the modified weights from Table 10. It can be observed from Figure 2 and Appendix Table A11, that driver ‘Eco design (D2)’ got top rank in sensitivity analysis almost all the time by varying weights of design criteria from 0.1 to 0.9. Similarly, driver ‘Incentives (MSH2)’ ranked last in all time in sensitivity analysis. Hence, it can be concluded from the sensitivity analysis that the findings achieved from BWM are free from any bias and are found to be stable.

Table 10. Benchmarking key results with previous studies

Research study	Technique used	Prioritised drivers/barriers
Mittal and Sangwan (2014)	Fuzzy TOPSIS	Future legislation, incentives and public pressure were the top three drivers of SM.
Ghazilla et al. (2015)	Delphi’s approach.	‘Increase in competition’ was the top priority driver. ‘Weak organization structures’ and ‘Inappropriate R&D facilities’ were top barriers of green manufacturing
Govindan, Diabat, and Shankar (2015)	Fuzzy-based MCDM approach	The top identified driver of green manufacturing was ‘Compliance with regulations’.
Agrawal and Vinodh (2019)	TISM	‘Flexibility in manufacturing’ was the most prominent factor affecting the sustainability of AM system.
Niaki, Torabi, and Nonino (2019)	Multi-stage survey	‘Customization’ and ‘Fabrication of complex design’ were the two top priority drivers for implementation of AM.

Figure 2. Drivers ranking by varying weights of criteria

5.2 Benchmarking key results with previous studies

The key results from previous studies is presented in Table 10.

6. Practical and managerial implications

The study facilitated industry practitioners to understand the significance and benefits of SAM towards the generation of light weight products with design freedom as well with lesser environmental impact. Also, there exist greater challenge to arrive at simplified product assemblies, enhanced product functionality with enhanced resource efficiency towards enhancement in operational efficiency with scope for remanufacturing and recycling. In this context, to successfully implement SAM, its key drivers to be analysed. In this viewpoint, this paper illustrates the analysis of drivers of SAM and the practitioners could focus on the prioritised drivers for enabling performance improvement. The study could facilitate managers for understanding the significance and motivation of SAM adaptation which will contribute towards deployment of SAM. The prioritisation of drivers would enable the deployment of relevant technologies and facilitate the identification of appropriate performance measures. The prioritised drivers would enable comprehensive understanding and assistance to industry practitioners on effective planning of strategies of SAM. The analysis and prioritisation of drivers would enable practicing managers to mitigate challenges in terms of sustainable design, restricted speed of AM, higher technology costs, compliance with standards and regulations. In a nutshell, the study would contribute towards setting a roadmap for researchers and practicing engineers to reinforce knowledge on SAM for feasible adoption in manufacturing domain for operational excellence.

7. Conclusions

AM processes are sustainable as there exist scope for reduced emissions and minimal environmental impacts. With increase in competition, people are more focusing on environmental related issues. They are showing increased concern towards environmental impacts associated with product. Sustainability has become a key concern for the present scenario. AM is one of the key technologies which has ability to meet sustainability requirement. In this context, SAM has good potential. In line with this, analysis of drivers of SAM is vital and hence in the present study, 40 drivers of SAM are being analysed. Data have been collected from four experts. The selected experts have high competence and experience in the field of SAM. BWM is used because it ensures more consistent result as benchmarked with other MCDM approaches (Ahmadi, Kusi-Sarpong, and Rezaei 2017). The top three key drivers include Eco design, green innovation, and energy conservation. The prioritisation of drivers of SAM helps the practitioners and industrial experts to develop efficient strategic plan for deploying SAM to minimise environmental impact. The conduct of the study facilitated the systematic analysis of SAM drivers.

8. Limitations and future research

In the present case study, 40 drivers are being included. In further studies, additional drivers could be considered based on managerial and technological advancements in manufacturing. The derived priority order of drivers could be further validated with other MCDM methods in future. Based on the priority order of drivers, appropriate techniques to be focused for deployment of SAM practices. Also, in future more expert opinion will be considered to enhance the accuracy and applicability of the result. In the future, aspects such as achievable component quality as well as the improved component function for SAM could be analysed.

Additional information

Notes on contributors

Rohit Agrawal

Rohit Agrawal is a Research Scholar in The Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India. He did his M.Tech. in Industrial Engineering and Management from the Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamilnadu, India. He completed his B.Tech. in Mechanical Engineering from Bhilai Institute of Technology, Durg, India. He has published various articles in reputed journals and international conference proceedings. He has skills and expertise of Multi-Criteria Decision Making, Fuzzy Theory, and Grey Theory. His areas of research are sustainability manufacturing, circular economy, additive manufacturing, Industry 4.0, and sustainable supply chain management.

S Vinodh

S.Vinodh is an Associate Professor in Production Engineering Department of National Institute of Technology, Tiruchirappalli, Tamil Nadu. He completed his Ph.D. from PSG College of Technology, Coimbatore, India. He completed his Master’s degree in Production Engineering from PSG College of Technology, Coimbatore, India and Bachelor’s degree in Mechanical Engineering from Government College of Technology, Coimbatore, India. He was a top ranker in his undergraduate and post graduate studies. He has been awarded National Doctoral Fellowship for pursuing research by AICTE, New Delhi. He has been awarded Outstanding Paper Award by Emerald publishers, UK for the year 2009 and 2018. He is the recipient of Institution of Engineers Young Engineer Award for the year 2013-14. He has published over 100 papers in International Journals. His research interests include Lean Production and Sustainable Manufacturing, Agile Manufacturing, Rapid Manufacturing, Product Development and Industry 4.0.

Appendix

Table A1 (a). Best and worst category specified by expert 1–4

Category	Identified as best by expert no.	Identified as worst by expert no.
Material	3
Management and stake holders		1
Customer/ Supplier/ Competitor		3
Collaboration and trends		2
Technology	2
Standards and regulations		4
Design	1
process	4

Table A1 (b). Comparison of best category with other categories (BO) by expert 1–4

Expert	Best to others	Material	Management and stake holders	Customer/ Supplier/ Competitor	Collaboration and trends	Technology	Standards and regulations	Design	process
E1	Design	3	9	7	4	3	7	1	3
E2	Technology	3	7	4	9	1	4	3	3
E3	Material	3	7	9	4	1	4	3	3
E4	process	3	7	7	4	3	9	3	1

Table A1 (c). Comparison of other categories with worst category (OW) by expert 1–4

Expert	E1	E2	E3	E4
Others to worst	Management and stake holders	Collaboration and trends	Customer/ Supplier/ Competitor	Standards and regulations
Material	7	7	7	7
Management and stake holders	1	3	3	3
Customer/ Supplier/ Competitor	3	5	1	3
Collaboration and trends	5	1	5	5
Technology	7	9	9	7
Standards and regulations	3	5	5	1
Design	9	7	7	7
process	7	7	7	9

Table A2 (a). Best and worst driver specified by expert 1–4 for material category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Optimum resources utilisation	1,4	2
Reusing and recycling materials and packaging	2	3
Less material wastage	3	1,4

Table A2 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for material category

Expert	Best to others	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage
E1	Optimum resources utilisation	1	3	7
E2	Reusing and recycling materials and packaging	7	1	3
E3	Less material wastage	5	7	1
E4	Optimum resources utilisation	1	4	9

Table A2 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for material category

Expert	E1	E2	E3	E4
Others to worst	Less material wastage	Optimum resources utilisation	Reusing and recycling materials and packaging	Less material wastage
Optimum resources utilisation	7	1	3	9
Reusing and recycling materials and packaging	5	7	1	5
Less material wastage	1	5	7	1

Table A3 (a). Best and worst driver specified by expert 1–4 for Management and stake holders category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Proper training and education	1	3
Incentives		1,4
Stake holder participation	3	2
Management support	4
Employee commitment towards sustainability	2

Table A3 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for management and stake holders category

Expert	Best to Others	Proper training and education	Incentives	Stake holder participation	Management support	Employee commitment towards sustainability
E1	Proper training and education	1	9	7	4	3
E2	Employee commitment towards sustainability	4	7	9	3	1
E3	Stake holder participation	9	7	1	3	3
E4	Management support	3	9	7	1	3

Table A3 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for management and stake holders category

Expert	E1	E2	E3	E4
Others to Worst	Incentives	Stake holder participation	Proper training and education	Incentives
Proper training and education	9	5	1	7
Incentives	1	3	3	1
Stake holder participation	3	1	9	3
Management support	5	7	7	9
Employee commitment towards sustainability	7	9	7	7

Table A4 (a). Best and worst driver specified by expert 1–4 for customer/ supplier/ competitor category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Customers’ expectations	1
Supplier commitment		2
Company image	2
Public pressure	4	1
Supply chain pressure		3,4
Competitor pressure towards greening	3

Table A4 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for customer/ supplier/ competitor category

Expert	Best to others	Customers’ expectations	Supplier commitment	Company image	Public pressure	Supply chain pressure	Competitor pressure towards greening
E1	Customers’ expectations	1	4	7	9	4	3
E2	Company image	3	9	1	7	4	3
E3	Competitor pressure towards greening	3	4	3	7	9	1
E4	Public pressure	3	3	3	1	9	3

Table A4 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for customer/ supplier/ competitor category

Expert	E1	E2	E3	E4
Others to worst	Public pressure	Supplier commitment	Supply chain pressure	Supply chain pressure
Customers’ expectations	9	7	7	7
Supplier commitment	5	1	5	7
Company image	3	9	7	7
Public pressure	1	3	3	9
Supply chain pressure	5	5	1	1
Competitor pressure towards greening	7	7	9	7

Table A5 (a). Best and worst driver specified by expert 1–4 for collaboration and trends category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Market trends		2,3
Economic benefit	2,4
Collaboration between organisations	1
Long term survival in market		4
Collaborative decision making	3	1

Table A5 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for collaboration and trends category

Expert	Best to others	Market trends	Economic benefit	Collaboration between organisations	Long term survival in market	Collaborative decision making
E1	Collaboration between organisations	4	3	1	3	9
E2	Economic benefit	9	1	4	4	3
E3	Collaborative decision making	9	3	3	4	1
E4	Economic benefit	3	1	3	9	4

Table A5 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for collaboration and trends category

Expert	E1	E2	E3	E4
Others to worst	Collaborative decision making	Market trends	Market trends	Long-term survival in market
Market trends	5	1	1	7
Economic benefit	7	9	7	9
Collaboration between organisations	9	5	7	7
Long term survival in market	7	5	5	1
Collaborative decision making	1	7	9	5

Table A6 (a). Best and worst driver specified by expert 1–4 for technology category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Green innovation	1,4
Energy conservation	3
Technology competence	2
Technology adaptability		4
Time to develop new product		2
Effective visual control		1,3

Table A6 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for technology category

Expert	Best to others	Green innovation	Energy conservation	Technology competence	Technology adaptability	Time to develop new product	Effective visual control
E1	Green innovation	1	3	4	3	7	9
E2	Technology competence	3	3	1	4	9	7
E3	Energy conservation	3	1	3	4	7	9
E4	Green innovation	1	3	7	9	4	7

Table A6 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for technology category

Expert	E1	E2	E3	E4
Others to worst	Effective visual control	Time to develop new product	Effective visual control	Technology adaptability
Green innovation	9	7	7	9
Energy conservation	7	7	9	7
Technology competence	5	9	7	3
Technology adaptability	7	5	5	1
Time to develop new product	3	1	3	5
Effective visual control	1	3	1	3

Table A7 (a). Best and worst driver specified by expert 1–4 for standards and regulations category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Environmental conservation	1,4
Emissions & global climate	3
Safety & security		1,2
Operational safety		3,4
Compliance with regulations	2

Table A7 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for standards and regulations category

Expert	Best to others	Environmental conservation	Emissions & global climate	Safety & security	Operational safety	Compliance with regulations
E1	Environmental conservation	1	3	9	7	4
E2	Compliance with regulations	3	3	9	7	1
E3	Emissions & global climate	3	1	5	7	3
E4	Environmental conservation	1	4	7	9	3

Table A7 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for standards and regulations category

Expert	E1	E2	E3	E4
Others to worst	Safety & security	Safety & security	Operational safety	Operational safety
Environmental conservation	9	7	5	9
Emissions & global climate	7	7	7	5
Safety & security	1	1	3	3
Operational safety	3	3	1	1
Compliance with regulations	5	9	5	7

Table A8 (a). Best and worst driver specified by expert 1–4 for design category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Light weight parts	2	3
Eco design	1,4
Flexibility in producing geometry	3	2,4
Ease of producing complex structure		1

Table A8 (b). Comparison of best driver with other drivers (BO) by expert 1–4 for design category

Expert	Best to others	Light weight parts	Eco design	Flexibility in producing geometry	Ease of of producing complex structure
E1	Eco design	3	1	7	9
E2	Light weight parts	1	3	9	7
E3	Flexibility in producing geometry	9	3	1	7
E4	Eco design	3	1	9	7

Table A8 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for design category

Expert	E1	E2	E3	E4
Others to worst	Ease of producing complex structure	Flexibility in producing geometry	Light weight parts	Flexibility in producing geometry
Light weight parts	7	9	1	7
Eco design	9	7	7	9
Flexibility in producing geometry	3	1	9	1
Ease of producing complex structure	1	3	3	3

Table A9 (a). Best and worst driver specified by expert 1–4 for process category

Driver	Identified as best by expert no.	Identified as worst by expert no.
Cost saving		3,4
Time saving	4
On demand manufacturing	2
Enables mass customisation	1
No specialised tooling required	3
Quality		1,2

Table A9 (b). Comparison of bbest driver with other drivers (BO) by expert 1–4 for process category

Expert	Best to others	Cost saving	Time saving	On demand manufacturing	Enables mass customisation	No specialised tooling required	Quality
E1	Enables mass customisation	4	3	3	1	4	9
E2	On demand manufacturing	4	4	1	4	3	9
E3	No specialised tooling required	9	4	3	3	1	7
E4	Time saving	9	1	3	3	4	7

Table A9 (c). Comparison of other drivers with worst driver (OW) by expert 1–4 for process category

Expert	E1	E2	E3	E4
Others to worst	Quality	Quality	Cost saving	Cost saving
Cost saving	5	5	1	1
Time saving	7	5	5	9
On demand manufacturing	7	9	7	7
Enables mass customisation	9	5	7	7
No specialised tooling required	5	7	9	5
Quality	1	1	3	3

Table A10. Weights of driver criteria for sensitivity analysis

Driver criteria	Normalised weight	Modified weights of driver criteria when design criteria varies from 0.1 to 0.9
		Run 1	Run 2	Run 3	Run 4	Run 4	Run 5	Run 6	Run 7	Run 8
Design (D)	0.1847	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9
Process (P)	0.1847	0.213	0.19	0.166	0.142	0.119	0.095	0.071	0.047	0.024
Material (M)	0.1331	0.154	0.137	0.12	0.103	0.085	0.068	0.051	0.034	0.017
Management and Stake Holders (MSH)	0.0502	0.058	0.052	0.045	0.039	0.032	0.026	0.019	0.013	0.006
Customer/ Supplier/ Competitor (CSC)	0.0611	0.071	0.063	0.055	0.047	0.039	0.031	0.024	0.016	0.008
Collaboration and Trends (CT)	0.0831	0.096	0.085	0.075	0.064	0.053	0.043	0.032	0.021	0.011
Technology (T)	0.231	0.131	0.116	0.102	0.087	0.073	0.058	0.044	0.029	0.015
Standards and Regulations (SR)	0.0713	0.178	0.158	0.138	0.118	0.099	0.079	0.059	0.039	0.02

Table A11. Ranking of drivers for sensitivity analysis for different weights

Driver Code	Normalised ranking	Run 1	Run 2	Run 3	Run 4	Run 4	Run 5	Run 6	Run 7	Run 8	Run 9
M1	4	1	3	4	4	5	5	5	5	5	1
M2	8	4	6	7	8	8	8	8	8	8	4
M3	12	10	13	13	14	14	14	14	14	14	10
MSH1	30	27	28	28	28	28	28	28	28	28	27
MSH2	40	40	40	40	40	40	40	40	40	40	40
MSH3	33	33	34	34	34	34	34	34	34	34	33
MSH4	28	25	26	26	26	26	26	26	26	26	25
MSH5	26	23	24	24	24	24	24	24	24	24	23
CSC1	24	21	22	22	22	22	22	22	22	22	21
CSC2	36	37	37	37	37	37	37	37	37	37	37
CSC3	29	26	27	27	27	27	27	27	27	27	26
CSC4	34	34	35	35	35	35	35	35	35	35	34
CSC5	38	39	39	39	39	39	39	39	39	39	39
CSC6	25	22	23	23	23	23	23	23	23	23	22
CT1	35	35	36	36	36	36	36	36	36	36	35
CT2	14	13	15	16	16	16	16	16	16	16	13
CT3	17	17	18	19	19	19	19	19	19	19	17
CT4	31	29	30	30	30	30	30	30	30	30	29
CT5	21	19	20	20	20	20	20	20	20	20	19
T1	2	12	14	14	15	15	15	15	15	15	12
T2	3	14	16	17	17	17	17	17	17	17	14
T3	6	16	17	18	18	18	18	18	18	18	16
T4	15	24	25	25	25	25	25	25	25	25	24
T5	18	32	33	33	33	33	33	33	33	33	32
T6	27	38	38	38	38	38	38	38	38	38	38
SR1	16	2	4	5	5	6	6	6	6	6	2
SR2	20	7	10	10	11	11	11	11	11	11	7
SR3	37	28	29	29	29	29	29	29	29	29	28
SR4	39	30	31	31	31	31	31	31	31	31	30
SR5	19	6	8	9	10	10	10	10	10	10	6
D1	5	15	2	2	2	2	2	2	2	2	15
D2	1	11	1	1	1	1	1	1	1	1	11
D3	13	18	9	3	3	3	3	3	3	3	18
D4	22	36	19	15	6	4	4	4	4	4	36
P1	23	20	21	21	21	21	21	21	21	21	20
P2	11	9	12	12	13	13	13	13	13	13	9
P3	7	3	5	6	7	7	7	7	7	7	3
P4	9	5	7	8	9	9	9	9	9	9	5
P5	10	8	11	11	12	12	12	12	12	12	8
P6	32	31	32	32	32	32	32	32	32	32	31