Analysis of barriers of sustainable 4D printing using Grey TOPSIS approach

ABSTRACT

The adoption of sustainable 4D printing in automotive component manufacturing industry helps to produce environmentally conscious critical components. This article introduces the study in which 18 sustainable 4D printing barriers were identified through expert consultation and literature review. The identified barriers were analysed against six criteria, namely, Build Time (C1), Accuracy (C2), Customization (C3), Carbon Footprint (C4), Material Utilization (C5) and Complexity (C6) using Grey Technique for Order of Preference by Similarity to Ideal Solution (G-TOPSIS) approach. Experts’ inputs were gathered to derive the relative importance of each barrier against each criterion in the form of Grey value. Using G-TOPSIS approach, the priority order of barriers is achieved. Based on the findings, significant barriers are found to be ‘Improper disposal strategy (B18)’, ‘Lack of interaction between smart materials and 3D printing technology (B8)’ and ‘Lack of Smart materials compatible with AM technologies (B10)’. The study facilitated industry practitioners to meticulously examine sustainable 4D printing barriers for improving production efficiency.

KEYWORDS:

• 4D printing
• sustainability
• barriers
• MCDM
• Grey TOPSIS

Introduction

Additive manufacturing integration with smart manufacturing is recognised for numerous reasons, such as real-time data analysis, creating complex shapes with good quality, decentralised production and augmented reality (Wankhede and Vinodh 2021). Moreover, it shows enormous capability in dealing with environmental impacts of new components (Laverne et al. 2019). 4D printing is the latest study topic that emerged with the introduction of 3D printing process. ‘4D printing is the development in the property, shape, and usefulness of a 3D printed structure over time when it is subjected to light, heat, pH, water, and so on’ (Choi et al. 2015; Kumar et al. 2021). As a new subject of study, 4D printing has various limits and obstacles that must be solved quickly (Kumar et al. 2021). The three primary hurdles of 4D printing are material, technological and design problems. Technological obstacles include the lack of 3D printing technology (Ahmed et al. 2021). Once these constraints are overcome, 4D printing will revolutionise product manufacture and design in the future. It will shorten marketing time, capital need, reduce the size of products, make items readily transportable, thereby creating an efficient and new business model (Ahmed et al. 2021; Azhar and Pei 2022).

3D printing technology was first discovered in 1987, and it has since attracted governments, researchers and enterprises to use it to produce distinct products from 3D models utilising computer-aided design (CAD) software (Nugroho et al. 2021). This initiative is prompted by many benefits based on 3D printing when compared to conventional production processes like injection moulding and extrusion. Such a method can construct a 3D structure with complicated geometry, which is extremely difficult to create with traditional injection moulding (Jian et al. 2022; Nugroho et al. 2021). Furthermore, it is centred on additive manufacturing technology, which constructs an item by applying raw materials layer by layer to reduce waste (Ameta et al. 2022; Maraveas, Bayer, and Bartzanas 2022). Subtractive manufacturing, for instance computer numerical control (CNC) machining, on the other hand, produces a component by reducing the quantity of material used. However, 3D printing cannot completely replace traditional production due to challenges with surface smoothness, printing speed and mechanical qualities of printed items. Because the fabrication process of this technique is slower than that of injection moulding and extrusion, it may not be suited for large-scale manufacturing (Mallakpour, Tabesh, and Hussain 2021). The generated items from 3D printing often have a rough surface due to the staircase effect (Nugroho et al. 2021; Trenfield et al. 2019). 3D printing with programming, stimuli, time and smart materials may be converted into 4D printing. The term time in this context does not refer to the time it takes to manufacture items. Skylar Tibbits demonstrated how alterations in a static produced item occur over time at the MIT TED conference in 2012. The term ‘electronic commerce’ implies to the auction of goods and services over the internet. This resulted in a new printing age encompassing a new dimension in 3D printing, namely, time, and this was given its own name, 4D printing technology. The fourth dimension is time, which can be inferred as 4D printing is time as an additional dimension of 3D printing (Demoly et al. 2021; Kumar et al. 2021; Mallakpour, Tabesh, and Hussain 2021).

4D printing possesses sustainable benefits such as efficient product design, efficient production operations, efficient supply chain, process optimisation and environmental-friendly manufacturing (Saritha and Boyina 2021). Efficient product design depicts product design to utilise less waste. Integration of physical and digital logistics through additive manufacturing drives sustainability in 4D printing. However, several barriers to sustainable additive manufacturing need to be recognised and analysed. In a nutshell, this study focussed on following research questions:

1. What are the barriers hindering sustainable 4D printing adoption in automotive industry?

2. How to recognise and evaluate the barriers?

In order to address the above research questions, following objectives have been framed:

1. To recognise potential sustainable 4D printing barriers that can help in realising adoption in automotive industry.

2. To derive priority order to identified barriers using MCDM approach in order to understand the step-by-step process of sustainable 4D printing adoption.

Literature review

The literature review with respect to 4D printing barriers and the related advancements is carried out as follows:

Choi et al. (2015) performed review on 4D printing technology. The authors discussed recent advances in fabrication process and 3D printed materials development enabling fabrication of smart materials. Moreover, the study compared 3D and 4D printing by identifying integration factors. Further, the authors discussed applications of 4D printing technology. The study revealed that these novel technologies helped in reducing production time and human efforts to assemble the machines or products. Further, 4D printed structures also aid in fulfilling the customer requirement of customised designs or products. The study also found that 4D printing enable fabrication of multilateral parts with desired functionalities and adaptability.

Azhar and Pei (2019) studied the relationship between 4D printing and manufacturing engineers and product designers through empirical studies. The study performed interviews and focussed group observation with six doctoral students of Brunel University. The research would help additive manufacturing and product engineers connect with each other in deploying 4D printing on parts.

González-Henríquez, Sarabia-Vallejos, and Rodriguez-Hernandez (2019) reviewed 4D printing technology discussing its basic principles, advantages, disadvantages, printing mechanism, and most suitable polymer materials. In addition to this, the study also presented the limitations, features and properties of existing polymer-based additive manufacturing systems. The authors stated that 4D printed parts derived from 3D printing that is capable to vary with time. The study’s findings delivered an overview of most widely deployed methodologies to manufacture polymer-based additive manufactured parts. Further, shape memory alloys, smart polymeric materials, active polymer-based composites and hydrogels were considered at the centre of 4D printing. In the future, novel advancements in terms of materials and machines were suggested concerning 4D printing. Also, such translation from 3D to 4D was considered as one of the challenges in manufacturing system.

Carlson and Li (2020) reported the kinetic assessment and development of low-cost-temperature-sensitive shape memory alloy for 4D printing. The study tested the filaments in fused deposition modelling process to develop 4D printed parts. The findings of the study revealed relationship amongst sample workpieces recovery rate, material concentration ratio and number of working cycles. Moreover, dynamic attributes such as angular velocity, angular acceleration, acceleration and recovery velocity were also examined to understand the shape memory polymer behaviour. The experimental analysis showed consistent fabricated filament with best shape memory attributes with 80% weight of PLA and 20% of TPU. The developed material also shown good shape recovery ratio and shape fixity. In future, similar research was suggested for other materials for 4D printing.

Liu and Chou (2020) carried out an extensive review on past work with respect to material fabrication and characterisation using deposition process along with discussion on influence of process parameters for textile applications. Further, several approaches in 4D printing for attaining shape morphing overtime were reviewed. The study also observed several strategies for creating shape memory polymer based preforms using 4D printing. Some of the significant challenges and opportunities in developing multidirectional composites for textile parts were provided. In future, development of continuous fibre reinforced revealing high shape memory performance was suggested for 4D printing of multidirectional preforms.

Saritha and Boyina (2021) executed a succinct literature review focusing on inception of 4D printing, applications, external stimuli response, conclusions and future opportunities. The authors mentioned 4D printing applications in tissue engineering, aerospace, organ engineering, medical technology and consumer products. The authors discussed two significant methods namely fused deposition modelling and stereolithography with regard to 4D printing. Further, 4D printing applications were shown in various fields such as clothing, automotive, healthcare, construction, defence and military.

Ahmed et al. (2021) discussed the basics and governing laws of 4D printing and materials used in it. The study also presented several applications like soft robotics and impediments that need to overcome for its smooth adoption in manufacturing field. Factors such as mathematical modelling, interaction mechanism, smart materials, additive manufacturing processes and stimuli were considered as responsible for 4D printing. The authors emphasised on smart materials applications as it responds to external stimulus having behaviours such as self-healing, self-capability, shape memory and self-assembly. In future, discovery of more materials was suggested to fully utilise 4D printing process.

Mallakpour, Tabesh, and Hussain (2021) discussed basic concept of 4D and 3D printing and their applications. The authors also mentioned about disadvantages of these technologies along with benefits. One disadvantage like low printing speed was mentioned that need to focus upon. The authors introduced 4D and 3D printing as novel technologies. The study shown significant applications of 4D printing in artificial organ production and other medical field. In future, discussion on challenges of 4D printing was recommended to explore.

Kumar et al. (2021) performed a review on 4D printing of biomedical applications. The authors stated the significance of 4D printing in changing dimensions over time which helps in fabricating customised products for ear tissue, bone and muscle development. The study suggested to recognise the drawbacks of these technologies for smooth adoption in future.

Demoly et al. (2021) aimed to emphasise potential barriers, status and challenges of 4D printing that need attention from a perspective of product-systems design. The authors made discussion on basics of smart materials, stimulus and additive manufacturing processes by highlighting the present status of 4D printing processes. The study further underlines challenges and opportunities of 4D printing from multiple viewpoints and developed a research framework for cross-disciplinary and engineering design. The findings of the study provided a research direction to the researchers for next decade towards smart product development.

Ameta et al. (2022) carried out systematic review on advancements of 4D printing and 3D printing technologies and made discussion on significance of these technologies in bringing sustainability and revolution with respect to industrial developments. The authors reviewed 4D printing applications from several fields including drug delivery. The study mentioned 4D printing advantages as environmental-friendly, customised products fabrication, cost-effective, reliable, speedy process and accuracy.

Azhar and Pei (2022) proposed theoretical framework for 4D printing by recognising the barriers of communication amongst engineers and designers in 4D printing parts design. In total, 60 designers and 50 engineers were involved to participate in the study. Findings of the study showed that 85% of participants were aware about 4D printing. Moreover, the study concludes that symbol analysis played significant role in ensuring easy understanding of 4D printing across all fields.

Based on the literature review, it can be seen that most of the studies critically reviewed 4D printing concept in order to derive challenges, current status, barriers and applications. The study conducted by Demoly et al. (2021) aimed to emphasise potential barriers, status and challenges of 4D printing. One more study conducted by Ahmed et al. (2021) discussed the basics and governing laws of 4D printing and materials used in it. Most of the studies focussed on applications of 4D printing in medical field. Moreover, Kumar et al. (2021) and Mallakpour, Tabesh, and Hussain (2021) suggested to explore the challenges of 4D printing for its smooth adoption. Ameta et al. (2022) discussed about 4D printing and sustainability but in the form of review in medical field. Therefore, very few studies had been reported the barriers of sustainable 4D printing in automotive industry. In this viewpoint, the present study dealt with identification and evaluation of sustainable 4D printing barriers in the context of automotive component manufacturing organisation using MCDM tool.

Methodology

The methodological procedure of study is shown in Figure 1.

Figure 1. Flow of methodology.

Sustainable 4D printing adoption in automotive component manufacturing industry aids in creating environmentally conscious critical components. In order to address the issue of adoption of sustainable 4D printing, the present study identified 18 barriers and analysed against six criteria namely Build Time (C1), Accuracy (C2), Customization (C3), Carbon Footprint (C4), Material Utilization (C5) and Complexity (C6). The selection of barriers and criteria was made through extensive literature review and conversations with eminent experts from Industry and academia. Moreover, expert’s consultation was taken to shortlist the identified barriers of sustainable 4D printing. After finalising the barriers, the input sheet was prepared to gather expert opinion in the form of linguistic variables. These linguistic variables are further transformed into Grey numbers for analysis. The solution methodology used is G-TOPSIS. Using this approach, grey separation distance is computed which helped in deriving closeness coefficient. Grey number-based closeness coefficient is then converted into crisp number and the ranking of barriers was derived.

A flow chart depicting the flow of methodology which initiates with literature review on barriers of sustainable 4D printing. Further, finalisation of barriers are done using experts opinion followed by analysis of identified barriers using Grey-TOPSIS approach. The stability of derived ranking is ensured through sensitivity analysis. Finally, results, discussions, implications and conclusions are derived.

Grey TOPSIS approach

Very few studies focused on analysing sustainable 4D printing barriers using MCDM approaches. Most popular MCDM approaches includes analytical hierarchical process, fuzzy-based decision-making techniques, data envelopment analysis, TOPSIS etc. These approaches have been integrated with fuzzy theory to reduce the inconsistencies in decision-making process. However, in the literature, Prakash et al. (2022), the authors utilised Grey-TOPSIS approach to capture the sustainable supply chain collaboration. Grey-based technique helped in overcoming ambiguity and vagueness in selection procedure and specifically during early stage of data gathering. Similar to other MCDM approaches, Grey TOPSIS approach also possesses few benefits and limitations. Benefits includes relative mathematical transparency and easy to use. It can be very operative in handling variety of factors. Moreover, Grey TOPSIS approach can provide effective solution with limited decision maker involvement. Acknowledging these benefits, the present study utilises Grey TOPSIS approach for analysis.

Theoretical background of Grey numbers

The Grey theory concept is based on Grey numbers (Deng 1989; Ju-Long 1982). Grey numbers can be utilised for several systems possessing incomplete, inadequate and discrete information with uncertainties (Bai and Sarkis 2018; Feng et al. 2019). Fundamentally, white system is with complete known information and black system with unknown information; however, if partial information is available with the system is termed as grey system (Badi, Bouraima, and Jibril 2022; Dey and Chakraborty 2016). Fundamental concept and definitions of Grey number could be referred from Dey and Chakraborty (2016).

Suppose $\otimes Gr{N}_1=\left[\underset{\_}{Gr{N}_1},\overline{Gr{N}_1}\right]$ and $\otimes Gr{N}_2=\left[\underset{\_}{Gr{N}_2},\overline{Gr{N}_2}\right]$be two grey numbers with $\mathrm{l}\left(\otimes Gr{N}_1\right)=\underset{\_}{Gr{N}_1}-\overline{Gr{N}_1}$, and $\mathrm{l}\left(\otimes \mathrm{G}\mathrm{r}{\mathrm{N}}_2\right)=\underset{\_}{\mathrm{G}\mathrm{r}{\mathrm{N}}_2}-\overline{\mathrm{G}\mathrm{r}{\mathrm{N}}_2}$.

The greater degrees for interval grey numbers are specified as:

(1) $R\left(\otimes Gr{N}_1\ge \otimes Gr{N}_2\right)=\{\begin{array}{c}1\\ \frac{\overline{Gr{N}_1}-Gr{N}_2}{1\left(\otimes Gr{N}_1\right)+1\left(\otimes Gr{N}_2\right)}\\ 0\end{array}\text{break}\begin{array}{c}\overline{Gr{N}_1}\ge \underset{\_}{Gr{N}_2}\\ \overline{Gr{N}_1}>Gr{N}_2\wedge Gr{N}_1<\overline{Gr{N}_2}\\ \overline{Gr{N}_2}\le \underset{\_}{Gr{N}_2}\end{array}\}$(1)

if $\mathrm{R}\left(\otimes \mathrm{G}\mathrm{r}{\mathrm{N}}_1\ge \hspace{0.17em}\otimes \mathrm{G}\mathrm{r}{\mathrm{N}}_2\right)>50\mathrm{\%}$, then the grey number $\otimes \mathrm{G}\mathrm{r}{\mathrm{N}}_1$ is greater than $\otimes \mathrm{G}\mathrm{r}{\mathrm{N}}_2$ (Bai and Sarkis 2018). These relationships aid in advancing TOPSIS approach.

The adjusted weight for each criterion C and expert E can be computed using below Equation 2 (Malaga and Vinodh 2021):

(2) $\widetilde{{\text{w}}_{\text{i}}^{\text{k}}}\text{=}\otimes {\text{w}}_{\text{i}}^{\text{k}}\times \otimes {\text{E}}^{\text{k}}\text{=}\left[\text{min}\left({\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}}\right),\text{max}\left({\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}}\right)\right]$(2)

The average adjusted criteria significance value is computed as:

(3) $\otimes \widetilde{{\mathrm{w}}_{\mathrm{i}}}=\frac{1}{\mathrm{k}}\left[\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^1+\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^2+\dots +\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^{\mathrm{k}}\right]$(3)

Equation 3 signifies the average adjusted criteria significance weight of i^th criteria of all experts. Further, for realising the effect of barriers against each criterion, the aggregated unadjusted grey values have been computed. Based on the number of decision makers E, grey values are computed as shown in Equation 4:

(4) $\otimes g_{ij}=\frac{1}{E}{\sum }_e^E\otimes {\mathrm{w}}_{\mathrm{i}\mathrm{j}}^e\times \otimes C^e,\mathrm{\forall }i\mathit{,}j$(4)

Further, adjusted aggregate grey values of barriers for each criterion are calculated using Equation 5 as:

(5) $\otimes {\tilde{g}}_{ij}=\otimes {\tilde{w}}_i\times \otimes g_{ij}=\left[\min \left({\tilde{w}}_i{g}_{ij},{\tilde{w}}_i{\overline{g}}_{ij},\overline{\tilde{w}}{g}_{ij},{\overline{\tilde{w}}}_i{g}_{ij}\right),\max \left({\underset{\_}{\tilde{w}}}_i-g_{ij},{\underset{\_}{\tilde{w}}}_i{\overline{g}}_{ij},\overline{\tilde{w}}-g_{ij},{\overline{\tilde{w}}}_i{\overline{g}}_{ij}\right)\right]\forall j\in e$(5)

TOPSIS approach

TOPSIS is applied to rank the identified barriers regarding adoption of sustainable 4D printing in automotive components manufacturing scenario. Various Multi-Criteria Decision Making (MCDM) techniques are established to deal with the unstructured problems containing several and contradictory goals, but TOPSIS approach is most widely used method. This method assumes two artificial hypotheses as follows:

H₁ Ideal alternative: hypothesis having the best level of all the barriers identified.

H₂ Negative ideal alternative: hypothesis having worst level of all barriers identified.

TOPSIS chooses the closest alternative to the positive ideal solution and furthest alternative from negative ideal solution.

The methodological stages of GTOPSIS are (Bai and Sarkis 2018; Dey and Chakraborty 2016):

1. To select set of criteria explaining alternatives.

2. To create decision matrix $\otimes \mathrm{G}$. In this matrix, $\otimes g_{ij}$ signify the grey assessment of i^th alternative with regards to j^th criterion.

3. Computing normalising decision matrix.

4. To compute criteria weights w_j.

5. To create the grey weighted normalised decision matrix.

6. Recognize the positive and negative ideal solutions.

   (6) $S^{+}=\left\{g_{\mathit{1}}^{+}\mathit{,}\dots {\mathit{.}\mathit{,}g}_n^{\mathit{+}}\right\}\mathit{=}\left\{\mathit{(}ma{x}_j{g}_{ij}\left|i\in P^{\prime }\mathit{)}\mathit{,}\mathit{(}\underset{j}{min}{g}_{ij}\right|i\in P^{\prime \prime }\right\}$(6)
   (7) $S^{\mathit{-}}\mathit{=}\left\{g_{\mathit{1}}^{-}\mathit{,}\dots {\mathit{.}\mathit{,}g}_n^{\mathit{-}}\right\}\mathit{=}\left\{\mathit{(}\underset{j}{min}{g}_{ij}\left|i\in P^{\prime }\mathit{)}\mathit{,}\mathit{(}\underset{j}{max}{g}_{ij}\right|i\in P^{\prime \prime }\right\}$(7)

where $P^{\prime }$ represents index value of highest Performance and $P^{\prime \prime }$ represents index value of lowest performance.

• (7) Computation of grey separation measures from positive and negative ideal solutions.

(8) ${P_j}^{+}\left(S^{+},S_j\right)={\sum }_{i=1}^n\otimes {\widetilde{g_i}}^{+}-\otimes \widetilde{g_i}={\sum }_{i=1}^n\left(\left(\underset{\_}{{\tilde{g}}_i^{+}}-{\bar{g}}_{ij}\right),\left({\tilde{\bar{g}}}_j^{+}-{\underset{\_}{g}}_{ij}\right)\right)$(8)

(9) $P_j^{-}\left(S^{-},S_j\right)={\sum }_{i=1}^n\otimes \widetilde{g_{ij}}-\otimes {\tilde{g}}_i^{-}={\sum }_{i=1}^n\left(\left(g_{ij}-{\bar{\tilde{g}}}_1\right),\left({\bar{g}}_{ij}-{\tilde{g}}_i^{-}\right)\right)$(9)

where $P_j^{+}$ represents positive ideal solution and $P_j^{-}$ represents negative ideal solution.

• (8) To compute relative closeness coefficient and derive ranking of the barriers.

(10) $\otimes {\mathrm{T}}_{\mathrm{i}}=\frac{\otimes {\mathrm{P}}_{\mathrm{j}}^{-}}{\otimes {\mathrm{P}}_{\mathrm{j}}^{+}}$(10)

where $T_i$ represents closeness coefficient.

Case study

Manufacturing organisations particularly in the automotive sector are expecting to develop sustainable products through additive manufacturing. The automotive organisations are moving towards embracing Industry 4.0 and planning for its smooth adoption in order to develop digital technological hub in organisation. In additive manufacturing, 4D printing is gaining much importance due to its ability of self-repair, self-assembly and multifunctionality. Moreover, organisations are struggling to improve production efficiency which can be achieved through adoption of 4D printing. However, adoption of sustainable 4D printing possess several barriers in organisation that need significant attention. In this viewpoint, the case organisation approached with a problem statement of recognising sustainable 4D printing barriers that can help industry in effortless implementation of it. Therefore, in line with organisation goals, an extensive literature was done to identify possible barriers of sustainable 4D printing and analysed it using MCDM approach. This work aims to examine identified sustainable 4D printing barriers using Grey TOPSIS approach and derive priority order to realise significant barrier.

Selection of sustainable 4D printing barriers and criteria

With the help of extensive literature review, possible barriers of sustainable 4D printing were identified related to automotive component organisation. From identified barriers, 18 barriers and 6 criteria were finalised in discussion with academia and industrial experts. Considering the expertise on 4D printing and sustainability, a group of three experts from automotive component manufacturing industry and academic institution is formed. The first expert was a strategic manager in the case automotive component organisation, whereas the second expert was selected from an associated automotive organisation working at a senior level position. Both the experts had rich expertise in additive manufacturing and aimed to focus on 4D printing process. The third expert was chosen from an academic institute serving at professor level position and possessing expertise in Industry 4.0 and additive manufacturing. The experts were serving at strategic managerial positions having more than 10 years of valuable experience 3D printing and manufacturing automation. The industry experts are responsible for carrying out tactical decision-making with respect to product innovation and fabrication. The academia expert performed several studies with respect to concept implementation and helped industries in smooth adoption of it. Further, a brainstorming session was done with experts to realise the present situation of the organisation. The experts utilised their rich experience in finalising the barriers of sustainable 4D printing. Table 1 depicts a description of sustainable 4D printing barriers finalised through expert opinion.

Table 1. Barriers of sustainable 4D printing and their description.

Sustainable 4D printing Barrier	Description	References
Lack of availability of materials that respond in real-time (B1)	This barrier refers to the lack of availability of sustainable smart materials that tends to respond in real time ensuring customization of products.	Ahmed et al. (2021); Demoly et al. (2021); Mallakpour, Tabesh, and Hussain (2021)
Inconvenience in creating complicated shapes (B2)	Difficulty in creating intricate shapes based on smart applications due to existing 3D printing systems.	Ahmed et al. (2021); Demoly et al. (2021); Mallakpour, Tabesh, and Hussain (2021)
Lack of awareness about 4D printing benefits (B3)	Less awareness of 4D printing advantages serves as barrier for sustainable 4D printing.	Ahmed et al. (2021); Demoly et al. (2021); Mallakpour, Tabesh, and Hussain (2021)
Synthetic versatility and adaptability of polymers (B4)	Adaptability of smart polymers with synthetic versatility is a barrier of sustainable 4D printing.	Ahmed et al. (2021); Choi et al. (2015); Kumar et al. (2021)
Requirement of additional developments to meet robust system (B5)	Additional advancements to existing 3D printing systems are required to create robust system. Non-availability of such system serves as barrier of sustainable 4D printing.	Ahmed et al. (2021)
Lack of dynamic response for structures and systems (B6)	Lack of dynamic response for structures and systems hinders adoption of sustainable 4D printing.	Ahmed et al. (2021); Choi et al. (2015); Kumar et al. (2021)
Degradation of smart materials with continuous deformation (B7)	Smart materials degradation needs to focus upon for smooth deployment of sustainable 4D printing.	Ahmed et al. (2021); Choi et al. (2015); Kumar et al. (2021)
Lack of interaction between smart materials and 3D printing technology (B8)	Interaction between 3D printing technology and smart materials is essential for 4D printing.	Ahmed et al. (2021)
Requirement of Designing of the smart structures (B9)	Smart structure design is required as it can sense external stimuli which forms basis of 4D printing.	Ahmed et al. (2021)
Lack of Smart materials compatibles with AM technologies (B10)	Compatible smart materials are essential for sustainable 4D printing.	Ahmed et al. (2021); Demoly et al. (2021); Mallakpour, Tabesh, and Hussain (2021)
Difficulty in inducing anisotropy by the layer-by-layer deposition mode (B11)	Anisotropy helps in bending 4D printed structures leading to responsive mechanism. Thus, difficulty in inducing anisotropy through layer-by-layer deposition is barrier for sustainable 4D printing.	Demoly et al. (2021)
Requirement of post-processing step (B12)	Necessity of post-processing is barrier for sustainable 4D printing.	Demoly et al. (2021)
Consolidating technical functions with multi-material and multi-functional objects (B13)	Integrating technical aspects with respect to multi-functional and multi-material objects serves barrier for 4D printing adoption.	Demoly et al. (2021)
Integrating multi-material functionality and technical with user-interface, user-centric, systems functions and user-experience (B14)	Combining multi-material functionality and technical with user-interface, user-centric, systems functions and user-experience is identified as barrier for sustainable 4D printing adoption.	Demoly et al. (2021)
Lack of design innovation with AM approaches for multi-material, transformation, and multifunctional skills through 4D printing (B15)	Design innovation with the help of additive manufacturing is necessary for sustainability of 4D printing.	Demoly et al. (2021)
Need of programming for materials responsive towards moisture due to slow reverse response (B16)	Requirement of programming helps in realizing the materials response for moisture. Embedding such programming in 4D printing becomes barrier for sustainability in 4D printing.	Demoly et al. (2021)
Environmental-friendly technologies (B17)	Lack of environmental-friendly technologies is the barrier for 4D printing ensuring sustainability.	Nugroho et al. (2021)
Improper disposal strategy (B18)	Although there is less waste in respect of materials in 4D printing, still improper disposal strategy hinders sustainability of it.	Nugroho et al. (2021)

The experts inputs are gathered for criteria and barriers analysis using linguistic variables and presented in Table 2.

Table 2. Experts inputs for criteria and barriers analysis.

	C1			C2			C3			C4			C5			C6
	E1	E2	E3	E1	E2	E3	E1	E2	E3	E1	E2	E3	E1	E2	E3	E1	E2	E3
	H	SH	H	SH	H	SH	SH	H	H	H	VH	H	H	H	VH	SH	H	H
B1	SG	G	SG	G	SG	SG	SG	F	SG	G	SG	G	G	SG	G	SG	F	SG
B2	SG	F	SG	G	F	F	F	SG	F	F	F	SG	F	SF	F	G	F	SG
B3	F	SG	F	F	G	G	SG	F	SG	F	SG	F	SG	F	SG	G	F	G
B4	SG	F	SG	G	F	SG	G	F	G	SG	F	SG	SG	F	SG	G	VG	G
B5	F	SG	F	F	SG	F	G	SG	F	F	SG	F	G	F	G	SG	F	SG
B6	G	F	G	SG	F	SG	F	SG	G	F	F	SG	F	SG	F	G	G	F
B7	SG	F	SG	F	SF	F	SG	F	SG	SG	F	SG	F	SG	F	SG	F	SG
B8	F	SG	F	G	F	G	SG	F	SG	F	G	G	G	VG	G	G	F	G
B9	SG	F	SG	G	F	SG	F	SG	SG	F	SF	F	F	SG	F	G	VG	G
B10	SG	F	G	F	SG	F	SG	F	SG	G	F	SG	VG	G	G	G	F	G
B11	G	F	G	F	SG	F	F	SG	SG	F	SG	F	G	F	F	G	F	G
B12	F	SF	F	F	SG	F	F	SG	F	F	SF	F	F	SF	F	G	F	F
B13	SF	F	SG	F	SG	F	SG	F	SG	G	F	SG	G	SG	G	G	F	SG
B14	SG	F	SG	F	SG	F	F	SG	F	SF	F	SF	F	SG	F	G	F	SG
B15	G	F	SG	F	SG	F	F	SG	F	F	F	SG	G	SG	G	G	F	SG
B16	SF	F	SG	F	SG	F	F	F	SG	SG	F	SG	G	F	G	F	SG	F
B17	F	SG	F	F	SG	G	F	SG	F	G	VG	G	G	G	G	F	F	SG
B18	SF	F	SG	F	SG	G	G	F	SG	G	VG	G	VG	G	G	F	SG	G

Note: *E1, E2, E3 - Expert.

Calculations using G-TOPSIS approach

The experts opinion of linguistic form has been transformed into grey numbers as shown in Tables 3 and 4 (Feng et al. 2019; Malaga and Vinodh 2021; Prakash et al. 2022). Moreover, the experience is also given weightage by considering its importance level as depicted in Table 5. In this study, based on the experience of experts, higher weightage is assigned to all experts, that is Very Important [0.9, 1.0] for the analysis.

Table 3. Criteria weights greyscale.

Scale level	Very High (VH)	High (H)	Somewhat High (SH)	Moderate (M)	Somewhat Moderate (SM)	Low (L)	Very (Low)
Weight ($\otimes w$)	[0.9,1.0]	[0.6,0.9]	[0.5,0.6]	[0.4,0.5]	[0.3,0.4]	[0.1,0.3]	[0,0.1]

Table 4. Greyscale for barriers rating.

Scale level	Very Good (VG)	Good (G)	Somewhat Good (SG)	Fair (F)	Somewhat Fair (SF)	Poor (P)	Very Poor (VP)
Weight ($\otimes g$)	[0.9,1.0]	[0.6,0.9]	[0.5,0.6]	[0.4,0.5]	[0.3,0.4]	[.1,.3]	[0,.1]

Table 5. Experts importance scale.

Scale level	Very Important	Important	Moderately Important
Weight ($\otimes w$)	[.9,1.0]	[.7,.9]	[.5,.7]

Calculation of criteria weight

The adjusted importance weight of each criterion is computed using Equation 2(2) $\widetilde{{\text{w}}_{\text{i}}^{\text{k}}}\text{=}\otimes {\text{w}}_{\text{i}}^{\text{k}}\times \otimes {\text{E}}^{\text{k}}\text{=}\left[\text{min}\left({\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}}\right),\text{max}\left({\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}},{\underset{\_}{\text{w}}}_{\text{i}}^{\text{k}}{\overline{\text{E}}}^{\text{k}},{\overline{W}}_{\text{i}}^{\text{k}}{\underset{\_}{\text{E}}}^{\text{k}}\right)\right]$(2) . Further, final average criteria weight is calculated using Equation 3(3) $\otimes \widetilde{{\mathrm{w}}_{\mathrm{i}}}=\frac{1}{\mathrm{k}}\left[\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^1+\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^2+\dots +\otimes {\widetilde{{\mathrm{w}}_{\mathrm{i}}}}^{\mathrm{k}}\right]$(3) as shown in Table 6.

Table 6. Adjusted importance weights of criteria.

Criteria	Weights
C1	(0.51,0.8)
C2	(0.48,0.7)
C3	(0.51,0.8)
C4	(0.63,0.93)
C5	(0.63,0.93)
C6	(0.51,0.8)

Calculation of unadjusted aggregate grey values of barriers on each criterion

Here, unadjusted aggregate grey values of barriers for each criterion are computed using Equation 4. The calculation of unadjusted aggregate grey value for Barrier 1 ‘Lack of availability of materials that respond in real-time’ with respect to Criteria 1 ‘Build Time’ is shown below:

(11) $\otimes g_{11}=\frac{1}{3}\left[\left(0.9\ast 0.5+0.9\ast 0.6+0.9\ast 0.5\right),\text{break}\left(1.0\ast 0.6+1.0\ast 0.9+1.0\ast 0.6\right)\right]=[0.48,0.7]$(11)

Similarly, unadjusted aggregate grey values for all barriers are computed and presented in Table 7.

Table 7. Unadjusted aggregate grey values of barriers with respect to criteria.

	C1	C2	C3	C4	C5	C6
B1	(0.48,0.7)	(0.48,0.7)	(0.42,0.567)	(0.51,0.8)	(0.51,0.8)	(0.42,0.567)
B2	(0.42,0.567)	(0.42,0.633)	(0.39,0.533)	(0.39,0.533)	(0.33,0.467)	(0.45,0.667)
B3	(0.39,0.534)	(0.48,0.767)	(0.42,0.567)	(0.39,0.533)	(0.42,0.567)	(0.48,0.767)
B4	(0.42,0.567)	(0.45,0.667)	(0.48,0.767)	(0.42,0.567)	(0.42,0.567)	(0.63,0.933)
B5	(0.39,0.534)	(0.39,0.533)	(0.45,0.667)	(0.39,0.533)	(0.48,0.767)	(0.42,0.567)
B6	(0.48,0.767)	(0.42,0.567)	(0.45,0.667)	(0.39,0.533)	(0.39,0.533)	(0.48,0.767)
B7	(0.42,0.567)	(0.33,0.467)	(0.42,0.567)	(0.42,0.567)	(0.39,0.533)	(0.42,0.567)
B8	(0.39,0.534)	(0.48,0.767)	(0.42,0.567)	(0.48,0.767)	(0.63,0.933)	(0.48,0.767)
B9	(0.42,0.567)	(0.45,0.667)	(0.42,0.567)	(0.33,0.467)	(0.39,0.533)	(0.63,0.933)
B10	(0.45,0.667)	(0.39,0.533)	(0.42,0.567)	(0.45,0.667)	(0.63,0.933)	(0.48,0.767)
B11	(0.48,0.767)	(0.39,0.533)	(0.42,0.567)	(0.39,0.533)	(0.42,0.633)	(0.48,0.767)
B12	(0.33,0.467)	(0.39,0.533)	(0.39,0.533)	(0.33,0.467)	(0.33,0.467)	(0.42,0.633)
B13	(0.36,0.5)	(0.39,0.533)	(0.42,0.567)	(0.45,0.667)	(0.51,0.8)	(0.45,0.667)
B14	(0.42,0.567)	(0.39,0.533)	(0.39,0.533)	(0.3,0.433)	(0.39,0.533)	(0.45,0.667)
B15	(0.45,0.667)	(0.39,0.533)	(0.39,0.533)	(0.39,0.533)	(0.51,0.8)	(0.45,0.667)
B16	(0.36,0.5)	(0.39,0.533)	(0.39,0.533)	(0.42,0.567)	(0.48,0.767)	(0.39,0.533)
B17	(0.39,0.534)	(0.45,0.667)	(0.39,0.533)	(0.63,0.933)	(0.54,0.9)	(0.39,0.533)
B18	(0.36,0.5)	(0.45,0.667)	(0.45,0.667)	(0.63,0.933)	(0.63,0.933)	(0.45,0.667)

Calculation of adjusted aggregate barriers grey values

Findings of Table 7 unadjusted grey values are further converted into adjusted aggregated grey values for all barriers using Equation 5(5) $\otimes {\tilde{g}}_{ij}=\otimes {\tilde{w}}_i\times \otimes g_{ij}=\left[\min \left({\tilde{w}}_i{g}_{ij},{\tilde{w}}_i{\overline{g}}_{ij},\overline{\tilde{w}}{g}_{ij},{\overline{\tilde{w}}}_i{g}_{ij}\right),\max \left({\underset{\_}{\tilde{w}}}_i-g_{ij},{\underset{\_}{\tilde{w}}}_i{\overline{g}}_{ij},\overline{\tilde{w}}-g_{ij},{\overline{\tilde{w}}}_i{\overline{g}}_{ij}\right)\right]\forall j\in e$(5) . Sample computation of Barrier 1 is shown below:

For Barrier 1, ‘Lack of availability of materials that respond in real-time’ with respect to Criterion 1 ‘Build Time’ is:

$\otimes {\tilde{g}}_{11}=\left[min\left(0.48\ast 0.51,0.48\ast 0.8,0.70\ast 0.51,0.70\ast 0.80\right),\text{breakmax}\left(0.48\ast 0.51,0.48\ast 0.8,0.70\ast 0.51,\text{break}0.70\ast 0.80\right)\left]=\right[0.245,0.56\right]$

The adjusted aggregated grey values of all barriers are shown in Table 8.

Table 8. Adjusted aggregated barriers grey values.

	C1	C2	C3	C4	C5	C6
B1	(0.245,0.56)	(0.23,0.49)	(0.214,0.454)	(0.321,0.746)	(0.321,0.746)	(0.214,0.454)
B2	(0.214,0.454)	(0.202,0.443)	(0.199,0.426)	(0.246,0.497)	(0.208,0.436)	(0.23,0.534)
B3	(0.199,0.427)	(0.23,0.537)	(0.214,0.454)	(0.246,0.497)	(0.265,0.529)	(0.245,0.614)
B4	(0.214,0.454)	(0.216,0.467)	(0.245,0.614)	(0.265,0.529)	(0.265,0.529)	(0.321,0.746)
B5	(0.199,0.427)	(0.187,0.373)	(0.23,0.534)	(0.246,0.497)	(0.302,0.716)	(0.214,0.454)
B6	(0.245,0.614)	(0.202,0.397)	(0.23,0.534)	(0.246,0.497)	(0.246,0.497)	(0.245,0.614)
B7	(0.214,0.454)	(0.158,0.327)	(0.214,0.454)	(0.265,0.529)	(0.246,0.497)	(0.214,0.454)
B8	(0.199,0.427)	(0.23,0.537)	(0.214,0.454)	(0.302,0.716)	(0.397,0.87)	(0.245,0.614)
B9	(0.214,0.454)	(0.216,0.467)	(0.214,0.454)	(0.208,0.436)	(0.246,0.497)	(0.321,0.746)
B10	(0.23,0.534)	(0.187,0.373)	(0.214,0.454)	(0.284,0.622)	(0.397,0.87)	(0.245,0.614)
B11	(0.245,0.614)	(0.187,0.373)	(0.214,0.454)	(0.246,0.497)	(0.265,0.591)	(0.245,0.614)
B12	(0.168,0.374)	(0.187,0.373)	(0.199,0.426)	(0.208,0.436)	(0.208,0.436)	(0.214,0.506)
B13	(0.184,0.4)	(0.187,0.373)	(0.214,0.454)	(0.284,0.622)	(0.321,0.746)	(0.23,0.534)
B14	(0.214,0.454)	(0.187,0.373)	(0.199,0.426)	(0.189,0.404)	(0.246,0.497)	(0.23,0.534)
B15	(0.23,0.534)	(0.187,0.373)	(0.199,0.426)	(0.246,0.497)	(0.321,0.746)	(0.23,0.534)
B16	(0.184,0.4)	(0.187,0.373)	(0.199,0.426)	(0.265,0.529)	(0.302,0.716)	(0.199,0.426)
B17	(0.199,0.427)	(0.216,0.467)	(0.199,0.426)	(0.397,0.87)	(0.34,0.84)	(0.199,0.426)
B18	(0.184,0.4)	(0.216,0.467)	(0.23,0.534)	(0.397,0.87)	(0.397,0.87)	(0.23,0.534)

Results and discussion

The present study intended to prioritise sustainable 4D printing barriers pertaining to automotive component manufacturing organisation using G-TOPSIS approach. Based on the G-TOPSIS analysis, Table 8 provides aggregated adjusted grey values of each barrier with respect to each criterion. In continuation of the same, Positive ideal and Negative ideal solutions for each criterion have been computed. The computation of positive and negative ideal solution has been done by referring to the Equations 6(6) $S^{+}=\left\{g_{\mathit{1}}^{+}\mathit{,}\dots {\mathit{.}\mathit{,}g}_n^{\mathit{+}}\right\}\mathit{=}\left\{\mathit{(}ma{x}_j{g}_{ij}\left|i\in P^{\prime }\mathit{)}\mathit{,}\mathit{(}\underset{j}{min}{g}_{ij}\right|i\in P^{\prime \prime }\right\}$(6) and 7(7) $S^{\mathit{-}}\mathit{=}\left\{g_{\mathit{1}}^{-}\mathit{,}\dots {\mathit{.}\mathit{,}g}_n^{\mathit{-}}\right\}\mathit{=}\left\{\mathit{(}\underset{j}{min}{g}_{ij}\left|i\in P^{\prime }\mathit{)}\mathit{,}\mathit{(}\underset{j}{max}{g}_{ij}\right|i\in P^{\prime \prime }\right\}$(7) . Table 9 shows Positive and Negative Ideal solutions for each criterion.

Table 9. Positive and negative ideal solutions of criteria.

Criteria	C1	C2	C3	C4	C5	C6
Positive Ideal	(0.613,0.613)	(0.536,0.536)	(0.613,0.613)	(0.870,0.870)	(0.870,0.870)	(0.746,0.746)
Negative Ideal	(0.168,0.168)	(0.158,0.158)	(0.198,0.198)	(0.189,0.189)	(0.207,0.207)	(0.198,0.198)

Calculation of Grey separation measure and closeness coefficient

Using positive and negative ideal solutions, grey separation distance from positive ideal and negative points for each barrier is computed. Equations 8(8) ${P_j}^{+}\left(S^{+},S_j\right)={\sum }_{i=1}^n\otimes {\widetilde{g_i}}^{+}-\otimes \widetilde{g_i}={\sum }_{i=1}^n\left(\left(\underset{\_}{{\tilde{g}}_i^{+}}-{\bar{g}}_{ij}\right),\left({\tilde{\bar{g}}}_j^{+}-{\underset{\_}{g}}_{ij}\right)\right)$(8) and 9(9) $P_j^{-}\left(S^{-},S_j\right)={\sum }_{i=1}^n\otimes \widetilde{g_{ij}}-\otimes {\tilde{g}}_i^{-}={\sum }_{i=1}^n\left(\left(g_{ij}-{\bar{\tilde{g}}}_1\right),\left({\bar{g}}_{ij}-{\tilde{g}}_i^{-}\right)\right)$(9) have been utilised to compute grey separation measure for each barrier.

Sample Calculation of grey separation measure for first barrier ‘Lack of availability of materials that respond in real-time (B1)’

$Positive:P_1^{+}=\left[\left(0.613-0.56\right),\left(0.613-0.245\right)\right]+{\sum }_{i=2}^n\otimes {\tilde{g}}_i^{+}-\otimes {\tilde{g}}_i=\left[0.0536,0.3688\right]+{\sum }_{i=2}^n\otimes {\tilde{g}}_i^{+}-\otimes {\tilde{g}}_i=\left(0.801,2.705\right)$

$Negative:P_1^{-}\left[\left(0.245-0.168\right),\left(0.560-0.168\right)\right]+\text{break}{\sum }_{i=2}^n\otimes {\tilde{g}}_{ij}-\otimes {\tilde{g}}_i^{-}=\left[0.0765,0.3917\right]+{\sum }_{i=2}^n\otimes {\tilde{g}}_{i}j-\otimes {\tilde{g}}_i^{-}=\left(0.425,2.329\right)$

Similarly, grey separation measures for all the barriers have been computed. Table 11 shows the grey separation measure of barriers.

Table 10. Closeness coefficient and Grey separation measure of barriers.

Barriers	Grey Separation Measure: Positive	Grey Separation Measure: Negative	Grey Closeness Coefficient
Lack of availability of materials that respond in real-time (B1)	(0.801,2.705)	(0.425,2.329)	(0.084,0.463)
Inconvenience in creating complicated shapes (B2)	(1.462,2.954)	(0.176,1.668)	(0.038,0.361)
Lack of awareness about 4D printing benefits (B3)	(1.194,2.853)	(0.277,1.936)	(0.058,0.404)
Synthetic versatility and adaptability of polymers (B4)	(0.913,2.726)	(0.404,2.217)	(0.082,0.449)
Requirement of additional developments to meet robust system (B5)	(1.251,2.874)	(0.257,1.879)	(0.054,0.395)
Lack of dynamic response for structures and systems (B6)	(1.099,2.839)	(0.291,2.031)	(0.06,0.417)
Degradation of smart materials with continuous deformation (B7)	(1.537,2.94)	(0.19,1.593)	(0.042,0.351)
Lack of interaction between smart materials and 3D printing technology (B8)	(0.634,2.664)	(0.466,2.496)	(0.09,0.484)
Requirement of Designing of the smart structures (B9)	(1.198,2.832)	(0.298,1.932)	(0.063,0.406)
Lack of Smart materials compatibles with AM technologies (B10)	(0.785,2.695)	(0.435,2.345)	(0.086,0.465)
Difficulty in inducing anisotropy by the layer-by-layer deposition mode (B11)	(1.11,2.85)	(0.28,2.02)	(0.057,0.415)
Requirement of post-processing step (B12)	(1.701,3.067)	(0.063,1.43)	(0.014,0.318)
Consolidating technical functions with multi-material and multi-functional objects (B13)	(1.122,2.832)	(0.298,2.008)	(0.062,0.415)
Integrating multi-material functionality and technical with user-interface, user-centric, systems functions and user-experience (B14)	(1.564,2.987)	(0.143,1.567)	(0.031,0.344)
Lack of design innovation with AM approaches for multi-material, transformation, and multifunctional skills through 4D printing (B15)	(1.141,2.839)	(0.291,1.989)	(0.06,0.412)
Need of programming for materials responsive towards moisture due to slow reverse response (B16)	(1.381,2.916)	(0.214,1.749)	(0.046,0.375)
Environmental-friendly technologies (B17)	(0.794,2.702)	(0.428,2.336)	(0.085,0.464)
Improper disposal strategy (B18)	(0.576,2.599)	(0.531,2.554)	(0.103,0.496)

Table 11. Final priority order of barriers.

Barriers	Rank
Lack of availability of materials that respond in real-time (B1)	V
Inconvenience in creating complicated shapes (B2)	XV
Lack of awareness about 4D printing benefits (B3)	XII
Synthetic versatility and adaptability of polymers (B4)	VI
Requirement of additional developments to meet robust system (B5)	XIII
Lack of dynamic response for structures and systems (B6)	VII
Degradation of smart materials with continuous deformation (B7)	XVI
Lack of interaction between smart materials and 3D printing technology (B8)	II
Requirement of Designing of the smart structures (B9)	XI
Lack of Smart materials compatibles with AM technologies (B10)	III
Difficulty in inducing anisotropy by the layer-by-layer deposition mode (B11)	IX
Requirement of post-processing step (B12)	XVIII
Consolidating technical functions with multi-material and multi-functional objects (B13)	VIII
Integrating multi-material functionality and technical with user-interface, user-centric, systems functions and user-experience (B14)	XVII
Lack of design innovation with AM approaches for multi-material, transformation, and multifunctional skills through 4D printing (B15)	X
Need of programming for materials responsive towards moisture due to slow reverse response (B16)	XIV
Environmental-friendly technologies (B17)	IV
Improper disposal strategy (B18)	I

Equation 10(10) $\otimes {\mathrm{T}}_{\mathrm{i}}=\frac{\otimes {\mathrm{P}}_{\mathrm{j}}^{-}}{\otimes {\mathrm{P}}_{\mathrm{j}}^{+}}$(10) has been utilised to compute Grey closeness coefficient.

Sample Calculation

For barrier ‘Lack of availability of materials that respond in real-time (B1)’:

Grey Closeness Coefficient $\otimes T_1$ = [$(\frac{0.425}{2.705+2.329}$), $(\frac{2.329}{2.705+2.329}$)] = (0.084,0.463)

Similarly, Grey closeness coefficient has been computed to derive the priority order of all the barriers as shown in Table 10.

Ranking of barriers

Since the closeness coefficient value is in grey form, it becomes difficult to rank the barriers. The grey numbers have been converted into crisp values and ranking of barriers is derived as depicted in Table 11.

Sample Calculation

For Barrier ‘B1’: Closeness Coefficient (Average) = (0.084 + 0.463)/2 = 0.2735.

In a similar way, average closeness coefficient value for all barriers is computed and ranking is derived.

With reference to Table 11, ranking order of the sustainable 4D printing barriers is ‘Improper disposal strategy (B18)’ > ‘Lack of interaction between smart materials and 3D printing technology (B8)’ > ‘Lack of Smart materials compatibles with AM technologies (B10)’ > ‘Environmental-friendly technologies (B17)’ > ‘Lack of availability of materials that respond in real-time (B1)’ > ‘Synthetic versatility and adaptability of polymers (B4)’ > ‘Lack of dynamic response for structures and systems (B6)’ > ‘Consolidating technical functions with multi-material and multi-functional objects (B13)’ > ‘Difficulty in inducing anisotropy by the layer-by-layer deposition mode (B11)’ > ‘Lack of design innovation with AM approaches for multi-material, transformation, and multifunctional skills through 4D printing (B15)’ > ‘Requirement of Designing of the smart structures (B9)’ > ‘Lack of awareness about 4D printing benefits (B3)’ > ‘Requirement of additional developments to meet robust system (B5)’ > ‘Need of programming for materials responsive towards moisture due to slow reverse response (B16)’ > ‘Inconvenience in creating complicated shapes (B2)’ > ‘Degradation of smart materials with continuous deformation (B7)’ > ‘Integrating multi-material functionality and technical with user-interface, user-centric, systems functions and user-experience (B14)’ > ‘Requirement of post-processing step (B12)’. ‘Improper disposal strategy (B18)’ is found to be most influential barriers in adoption of sustainable 4D printing.

Sensitivity analysis

Sensitivity analysis aids in stabilising the identified ranking of alternatives. The reason being performing sensitivity analysis is variation in expert opinion on criteria. Therefore, in order to address this issue of variation, several scenarios, that is experiments need to be performed by interchanging weights of criteria (Bai and Sarkis 2018; Malaga and Vinodh 2021). Table 12 shows the interchanged weights considered for sensitivity analysis. Using these weights, the ranking of barriers for each experiment obtained is depicted in Table 13.

Table 12. Experiment-wise weights for each criterion.

Criteria	C1	C2	C3	C4	C5	C6
W1	(0.48, 0.7)	0.51, 0.8)	(0.51, 0.8)	(0.63, 0.933)	(0.63, 0.933)	(0.51, 0.8)
W2	(0.51, 0.8)	(0.51, 0.8)	(0.48, 0.7)	(0.63, 0.933)	(0.63, 0.933)	(0.51, 0.8)
W3	(0.51, 0.8)	(0.63, 0.933)	(0.51, 0.8)	(0.48, 0.7)	(0.63, 0.933)	(0.51, 0.8)
W4	(0.51, 0.8)	(0.48, 0.7)	(0.51, 0.8)	(0.63, 0.933)	(0.48, 0.7)	(0.51, 0.8)
W5	(0.51, 0.8)	(0.48, 0.7)	(0.51, 0.8)	(0.63, 0.933)	(0.63, 0.933)	(0.63, 0.933)
W6	(0.51, 0.8)	(0.48, 0.7)	(0.51, 0.8)	(0.63, 0.933)	(0.63, 0.933)	(0.48, 0.7)

Table 13. Sensitivity analysis.

Barriers	Present case	Exp.1	Exp.2	Exp.3	Exp.4	Exp.5	Exp.6
Lack of availability of materials that respond in real-time (B1)	5	5	5	4	3	4	5
Inconvenience in creating complicated shapes (B2)	15	15	15	15	15	15	15
Lack of awareness about 4D printing benefits (B3)	12	12	12	9	11	12	12
Synthetic versatility and adaptability of polymers (B4)	6	6	6	6	6	6	6
Requirement of additional developments to meet robust system (B5)	13	13	13	13	13	13	13
Lack of dynamic response for structures and systems (B6)	7	8	8	7	7	7	8
Degradation of smart materials with continuous deformation (B7)	16	16	16	16	16	16	16
Lack of interaction between smart materials and 3D printing technology (B8)	2	2	2	2	2	2	2
Requirement of Designing of the smart structures (B9)	11	9	11	7	8	7	11
Lack of Smart materials compatibles with AM technologies (B10)	3	3	3	3	5	3	3
Anisotropy induced by the layer-by-layer deposition mode (B11)	9	11	9	9	8	10	10
Requirement of post-processing step (B12)	18	18	18	18	18	18	18
Consolidating technical functions with multi-material and multi-functional objects (B13)	7	7	7	12	10	9	7
Integrating technical and multi-material functionality with user-centric, user-interface, user-experience, and systems functions (B14)	17	17	17	17	17	17	17
Lack of design innovation with AM approaches for transformation, multi-material, and multifunctional capabilities through 4D printing (B15)	9	9	9	9	11	11	9
Need of programming for materials responsive towards moisture due to slow reverse response (B16)	14	14	14	14	14	14	14
Environmental-friendly technologies (B17)	4	3	4	5	4	4	3
Improper disposal strategy (B18)	1	1	1	1	1	1	1

A clustered column chart of average closeness coefficient of each barrier with respect to each experiment performed for sensitivity analysis. Clustered columns show different experiments ensuring stability of ranking of sustainable 4D printing barriers.

Sensitivity analysis has been carried out to understand the ranking stability of obtained results. Sensitivity analysis ranking stability has been shown in Figure 2. In total, six experiments (Exp.) were carried out. In each experiment, criteria weights were interchanged, and was obtained. It can be clearly seen from Figure 2 that ranking stability of barriers obtained from average closeness coefficient is achieved. Thus, the findings of the study have been proved to be stable.

Figure 2. Sensitivity analysis.

Implications

The present study analysed sustainable 4D printing adoption barriers in Indian automotive component organisation using G-TOPSIS technique. Eighteen potential barriers hindering 4D printing adoption in organisation are recognised in discussion with eminent experts and extensive literature review. Six criteria were used for the analysis and derived priority order of barriers. The outcomes from the study recommended to concentrate on disposal strategy with regard to 3D printing ensuring sustainability. 4D printing produces less waste compared to 3D printing as minor adjustments can be immediately made in software to avoid the printing issues. However, organisations need to focus on appropriate disposal strategy for adoption of 4D printing. The study also provides enough understanding on 4D printing and sustainability that would help industry practitioners to create awareness about it. The study further recommends developing sufficient finance for adoption of 4D printing in order to achieve its long-term benefits. The barriers ‘Lack of interaction between smart materials and 3D printing technology (B8)’, ‘Lack of Smart materials compatibles with AM technologies (B10)’, Environmental-friendly technologies (B17)’ and ‘Lack of availability of materials that respond in real-time (B1) also need focus for taking advantage of 4D printing. Moreover, automotive manufacturing can develop short-term strategy to transform towards 4D printing using given priority order of barriers. Industry practitioners recognised the significance of the present approach of deriving ranking of barriers and taken necessary steps towards implementation of it.

Conclusions

Sustainability is gaining much attention over the last decade. It delivers understanding on energy and resource consumption which in turn reduces material quantity requirement in supply chain and ensures sustainable practices in manufacturing processes. The present study aimed to examine barriers of sustainable 4D printing and analysed using Grey TOPSIS approach. In total,18 barriers were identified through literature review and expert consultation and analysed against six criteria. Examination of each barrier was done through several rounds of discussion with the experts and gathered their opinion as per Grey TOPSIS methodology. Post data collection, Grey TOPSIS computations was done and derived priority order. The study outcomes showed that barrier ‘Improper disposal strategy (B18)’ as the most influential one that need to overcome for smooth adoption of sustainable 4D printing followed by ‘Lack of interaction between smart materials and 3D printing technology (B8)’, ‘Lack of Smart materials compatibles with AM technologies (B10)’, ‘Environmental-friendly technologies (B17)’ and ‘Lack of availability of materials that respond in real-time (B1)’. The study delivered sufficient understanding on 4D printing and sustainability that would help industry practitioners to build awareness about it.

Findings from the study steered the industry practitioners to take suitable measures for adoption of 4D printing. This study holds certain limitations that may act as opportunities for conducting future research. The study identified 18 sustainable 4D printing barriers and six criteria with respect to Indian automotive manufacturing organisation; in future, additional barriers could be included in line with technological advancements of additive manufacturing. Also, the life cycle assessment studies of as built product of 4D printing could be done. The present study deployed Grey TOPSIS approach for evaluation of barriers; in future integrated MCDM approaches could be used with additional barriers and criteria for deeper realisation of 4D printing adoption strategy. Moreover, future researchers may focus on performing comparative analysis of this study with other MCDM approaches. Since, the present study focusses on sustainable 4D printing barriers, future researchers may attempt to develop sustainable guidelines for 4D printing component and derive priority order of such guidelines. The structural model could also be developed for sustainable 4D printing barriers depicting hierarchical adoption strategy for industry. Furthermore, 4D printing barriers could be mapped with circular economy strategies and sustainable framework could be developed.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Vishal Ashok Wankhede

Vishal Ashok Wankhede is an Assistant Professor in the Department of Operations Management and Quantitative Techniques, Indian Institute of Management, Bodhgaya since July 2023. Prior to joining IIM Bodhgaya, he worked as an Assistant Professor in Department of Mechanical Engineering, School of Technology, PDEU Gandhinagar from March 2022 to June 2023. He also holds three years of teaching experience as an Assistant Professor in Department of Industrial Engineering, Pandit Deendayal Energy University, Gandhinagar from July 2016 – June 2019. He received his Ph.D.in the area of Industry 4.0 from Department of Production Engineering, National Institute of Technology, Tiruchirappalli in July 2022. He received his M.Tech in the area of Cloud Manufacturing from Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamilnadu in 2016. He received his B.Tech in Mechanical Engineering from Dr. Babasaheb Ambedkar Technological University, Lonere, Raigad, Maharashtra (Govt. Autonomous), Maharashtra in 2014. He has also received “Budding Researcher Award” on Institute Day 2022 of NIT Tiruchirappalli for the outstanding research performance during Ph.D. (July 2019 – March 2022) at Department of Production Engineering, NIT Tiruchirappalli. He is actively associated with various professional bodies such as Institution of Engineers (India), Indian Society for Technical Education, American Society of Mechanical Engineers and Indian Institution of Industrial Engineering. His research area includes Industry 4.0, Smart Manufacturing, Additive manufacturing, Circular economy, Sustainable Manufacturing etc.

S. Vinodh

S. Vinodh is a Professor in the Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India. He completed his PhD degree under AICTE National Doctoral Fellowship Scheme from PSG College of Technology, Coimbatore, India. He completed his Master’s in Production Engineering from PSG College of Technology, Coimbatore, India and Bachelor’s in Mechanical Engineering from Government College of Technology, Coimbatore, India. He has published over 175 papers in international journals. He received Highly Commended Paper Award from Emerald Publishers for the year 2016. He is the recipient of Innovative Student Project Award 2010 based on his PhD thesis from Indian National Academy of Engineering (INAE), New Delhi, India. His research interests include sustainable manufacturing, lean manufacturing, agile manufacturing, rapid manufacturing, product development and Industry 4.0.