Although there are many studies in the literature addressing digital transformation in the supply chain, there are no studies evaluating its strategies with a MCDM. Therefore, in this section, we reviewed the studies of digital transformation in supply chain to form the basis for our study. Xu (2014) gave important information about digital enterprise management required by decision-makers and managers in the organizations by focusing on digital enterprise and its managing. He also reviewed emerging trends and future directions, issues, and success factors of managing DSC. Uhl and Gollenia (2016) reviewed the combination of transformational capabilities and new digital skills to be developed. They also presented examples of a Digital Transformation Roadmap by introducing a set of different digital use cases related to supply chain management. Büyüközkan and Göçer (2018a) reviewed the state-of-the-art of the current DSC literature, detailing it from both academic and industrial perspectives. They also presented the main limitations and prospects in DSC, advantages, weaknesses, and limitations of individual methods. Büyüközkan and Göçer (2018b) proposed a new MCDM approach to evaluate the supplier selection process in the DSC environment. They presented a new framework that combines the interval-valued intuitionistic fuzzy (IVIF) AHP method to evaluate criterion weights and the Additive Ratio Assessment (ARAS) methodology to evaluate alternatives. Büyüközkan and Göçer (2019) used an approach that integrates PFSs into alternative DSC partner selection. Bienhaus and Haddud (2018) aimed to identify the effect of digitization on procurement and its role in the area of supply chain management. In the study, they also introduced potential obstacles to digitizing procurement and supply chains and ways to overcome them. Farahani et al. (2016a) presented an overview of the DSC management practices of leading companies in various industries, the DSC management concepts, and opportunities that arise from the application of digital technologies to supply chain management (SCM). Agrawal and Narain (2018) referred to its benefits by offering a framework of the digital supply chain. Scuotto et al. (2017) explained the relationship between multiple buyers and suppliers in the context of SMEs’ DSC management. Farahani et al. (2016b) provided the creation of the DSC management agenda by presenting 17 DSC management use cases identified by expert interviews. Korpela et al. (2016) aimed to establish a DSC integration based on global standards. Bhargava et al. (2013) proposed a new based approach for protecting shared data in DSCs. Pundir et al. (2019) reviewed the suitability of complementary technologies such as IoT and Blockchain technology for DSC. Luthra and Mangla (2018) evaluated challenges to Industry 4.0 initiatives for supply chain sustainability in developing economies using an extensive literature review. Büyüközkan and Göçer (2017) presented an approach evaluating with intuitionistic fuzzy sets the supplier selection process in the DSC environment. Using the MOORA (Multi-Objective Optimization with Ratio Analysis) method, they realized a real case study to show the validity of the proposed approach. Alkan (2021) used the interval-valued Pythagorean fuzzy AHP method to assess the risks of digital transformation based on a sustainable supply chain. Tjahjono et al. (2017) purposed to provide a thought towards Supply Chain 4.0 by presenting a preliminary analysis of the impact of Industry 4.0 on SCM. Ivanov et al. (2019) reviewed how digital technologies and Industry 4.0 affect the ripple effect and performance of the supply chains. They presented the first study that connects information, business, analytics, engineering, and perspectives on digitalization and supply chain risks.

Digital Supply Chain transformation is based on the full implementation of various new digital technologies. With the developing technologies, consumers, employees and business partners have more expectations, leading companies to develop more reliable and sensitive supply chains. Therefore, organizations need to adopt new technologies such as cloud, big data analytics, augmented reality, internet of things and 3D printing to keep up with digital transformation. These technologies enable the digitization of products and services, new business models and the digitization and integration of every link in an organization’s value chain (i.e. engineering and manufacturing, product development and innovation, digital workplace, distribution and digital sales channels and customer relationship management) will also offer enormous benefits through making production more responsive to consumer demand, reducing costs, saving consumers’ time and boosting employment (PwC Sweden, 2018; WTO, 2019). The faster these technologies develop in performance and cost, the faster they will make a change in SCM and will have a considerable impact on current and future SCM tasks (Kearney, 2015). The aim of the DSC is to completely integrate and make visible every aspect of the movement of goods and services. The most important technology that will fulfill this purpose of DSC is big data (PwC Sweden, 2018). Big data is considered as high velocity, high volume, and high variety information assets that demand cost-effective, innovative forms of information processing for decision making (Farahani et al., 2020). Big data in DSC is to realize the necessary transparency by uncovering process interruptions and ensuring that changes are implemented quickly. Big data analytics provide better demand forecasting and planning, inventory planning and management, network, and routing optimization advanced procurement with collaborative optimization (Kearney, 2015; Alkan and Kahraman, 2021). Cloud computing is described as a style of computing in which scalable and flexible IT-enabled capabilities are presented as a service through internet technologies (Farahani et al., 2020). Cloud computing creates diverse business networks to enable companies to fully and rapidly engage with supply chain stakeholders (Kearney, 2015). The Internet of Things (IoT) is a network of physical objects that includes embedded technology to communicate, perceive, or interact with their internal and external environments (Farahani et al., 2020). IoT provides to open up to new business models and operational possibilities in the supply chains and respond to changing customer needs in real-time effectively. Tracking and tracing throughout the supply chain are provided through technologies underlying IoT such as Bluetooth, GSM (global system for mobile communication), and radio frequency identification (RFID) to rapidly evaluate and respond to changes in customer demand (WTO, 2019). Warehouse automation through advanced robotic technologies becomes much more holistic as some warehouses are fully connected to production loading points, so that all processes are carried out without manual intervention (Alicke et al., 2016). A three-dimensional scanner (3D) is a device creating object models of them by capturing data about the appearance and shape of real-world objects (Farahani et al., 2020). With 3D printing in the supply chain, the spare parts supply chain can be decreased to much fewer suppliers, even making own production possible. Thereby, 3D having an important impact on physical flows in the supply chain leads to faster delivery to the customer, lower labour unit and transport cost, and notably reduced inventory levels and costs in the supply chain (PwC Sweden, 2018). Augmented reality is defined as the situation that creates a new perception environment by combining computer-generated elements with the real world, in which users can interact (WTO, 2019). Augmented reality in the supply chain contributes to finding the right quantity of the right material much more efficiently by enabling better warehouse management (Kearney, 2015). Except for these technologies, GPS technology allows companies to take full control of shipping locations, while sensors control environmental conditions such as temperature and humidity and determine maintenance requirements (PwC Sweden, 2018). Autonomous and smart vehicles provide significant operational cost reductions in transportation and product handling, and also offer several benefits related to lower environmental costs and lead times (Alicke et al., 2016).

Supply chain managers who want to implement digital transformation in their SCMs ensure that they not only identify the challenges and opportunities their organizations face, but also consider the way suppliers, customers and other market partners interact with their organizations by enabling the digital transformation of the entire organization, its services and products (Kearney, 2015). In SCM, organizations should apply various steps that are necessary to deliver a product or service to customers. According to the supply chain council, these steps can be operated with the help of the SCOR model which includes the Plan, Source, Make, Deliver, Return processes (Büyüközkan and Göçer, 2018a). This model includes all processes that meet lower costs and faster customer demand, helping to support communication between supply chain partners and increase the efficiency of supply chain management (Uhl and Gollenia, 2016). Each of these elements is quickly being digitalized through technological innovation and thereby the chain becomes an integrated system working flawlessly. As a result, a digital supply chain strategy must consider the issues and success factors of digital transformation in the supply chain and examine it as a holistic approach to reap the full benefits of digitalization.

Organizations seeking to establish a DSC will face competitive extinction unless they develop clear strategies that respond to the opportunities presented in an all-digital environment. An organization that wants to generate and measure long-term value should integrate its digital initiatives into its overall supply chain strategy. Therefore, a digital supply chain strategy should be an integral part of a company’s overall business model and organizational structure (Raab and Griffin-Cryan, 2011). Once the strategies are determined, companies must implement the DSC opportunities needed to carry out the transformation in their organizations. To ensure the effectiveness and efficiency of supply chains, the digital supply chain must be more agile and stronger by having the right people and skills, processes and tools in the right places. To achieve these goals, organizations need to work on such initiatives as focusing on better system and process standardization, create new business models, reconfigure demand forecasts for better interaction with the customer, enhancing sourcing capabilities in emerging markets and institutionalizing staff development better (Xu, 2014).

Developing strategies based on demand, people, technology, and new business models provide that all sections of the organizations fulfill the required changes to become more demand-driven, customer-focused, technology-savvy, and risk compliant. Organizations must develop demand-based strategies for digital transformation in supply chains. Through demand-based strategies, organizations can obtain real-time data by continuously communicating with customers. Human resources and skills-based strategies enable the development of people with various skills to achieve DSC results. Organizations should find people capable of collecting and analysing data to make better decisions and thus provide more customer-oriented growth solutions. IT and technology-based strategies help efficiently deploy knowledge and integrate information and communication technologies. New business models-based strategies enable changes in current business models of organizations for better customer interaction (Bailey et al., 2017).

Intuitionistic fuzzy sets proposed by Atannasov (1986) are an extension of the traditional fuzzy set theory. An IFS is defined by two membership values named as membership and non-membership that their sum is one or less than one. Definition 3.1.

Let X be a non-empty set. An IFS I in X is given by: (1) I = { ( x , μ I ( x ) , ν I ( x ) ) | x ϵ X } , where the function μ I : X → [ 0 , 1 ] and υ I : X → [ 0 , 1 ] defines the membership and non-membership degrees of an element to the sets I with the condition that (2) 0 ⩽ μ I ( x ) + ν I ( x ) ⩽ 1 , for ∀ x ϵ X . The hesitancy degree is calculated as follows: (3) π I ( x ) = 1 − μ I ( x ) − ν I ( x ) .

Pythagorean fuzzy sets (PFS) introduced as an extension of intuitionistic fuzzy set by Yager (2013) are defined two membership values named as membership and non-membership. In PFSs, the sum of membership and non-membership degrees assigned by decision-makers can exceed 1, but the sum of their squares must be at most 1. PFSs are defined in Definition 3.3. Definition 3.3.

Let X be a non-empty set. A Pythagorean fuzzy set P in X is an object having the form (Zhang and Xu, 2014): (6) P = { ⟨ x , μ P ( x ) , ν P ( x ) ⟩ | x ϵ X } , where the function μ P : X → [ 0 , 1 ] defines the degree of membership and υ P : X → [ 0 , 1 ] defines the degree of non-membership of the element x ∈ X to P and it holds that: (7) 0 ⩽ ( μ P ( x ) ) 2 + ( ν P ( x ) ) 2 ⩽ 1 , for ∀ x ϵ X . The hesitancy degree is calculated as follows: (8) π P ( X ) = 1 − μ P ( x ) 2 − ν P ( x ) 2 .

Yager (2017) introduced q-rung orthopair fuzzy sets, a general class of IFSs and PFSs. The sum of the qth power of membership and non-membersip degrees q-rung orthopair fuzzy sets is bounded with one. When q = 3, Senapati and Yager (2020) have called q-rung orthopair fuzzy sets as fermatean fuzzy sets (see Fig. 2).

In this section, the mathematical operations of IVFFSs have been briefly presented (Jeevaraj, 2021). Definition 3.10.

Let X be a fixed set. An IVFFSs F ˜ in X is an object having the form (22) F ˜ = { ⟨ x , μ F ˜ ( x ) , ν F ˜ ( x ) ⟩ | x ϵ X } , where μ F ˜ ( x ) ⊆ [ 0 , 1 ] and υ F ˜ ( x ) ⊆ [ 0 , 1 ] indicate the membership and non-membership degrees of the element x ∈ X to the set F ˜, respectively. Also, for each x ∈ X, μ F ˜ ( X ) and υ F ˜ ( X ) are closed intervals and their lower and upper bounds are denoted by μ F ˜ L ( x ), μ F ˜ U ( x ), υ F ˜ L ( x ), υ F ˜ U ( x ), respectively. Therefore, F ˜ can also be expressed as follows: (23) μ F ˜ ( x ) = [ μ F ˜ L ( x ) , μ F ˜ U ( x ) ] ⊆ [ 0 , 1 ] , (24) ν F ˜ ( x ) = [ υ F ˜ L ( x ) , υ F ˜ U ( x ) ] ⊆ [ 0 , 1 ] , where the expression is subject to the condition (25) 0 ⩽ ( μ F ˜ U ( x ) ) 3 + ( υ F ˜ U ( x ) ) 3 ⩽ 1 . For every x ∈ X, π F ˜ ( x ) = [ π F ˜ L ( x ) , π F ˜ U ( x ) ] is called the hesitancy degree in IVFFSs, where π F ˜ L ( x ) = 1 − ( μ F ˜ U ( x ) ) 3 − ( υ F ˜ U ( x ) ) 3 3 and π F ˜ U ( x ) = 1 − ( μ F ˜ L ( x ) ) 3 − ( υ F ˜ L ( x ) ) 3 3 .

Fermatean fuzzy sets, which are the extension of ordinary fuzzy sets, have been introduced by Senapati and Yager (2020). No study integrating FFSs with the AHP method has been performed in the literature. The steps of the proposed IVFF-AHP method whose flow chart is illustrated in Fig. 4 are given as follows:

Step 1: Construct the hierarchical structure by determining the criteria and alternatives.

Determine objective, decision criteria, and alternatives for the given problem. The set A i = { A 1 , A 2 , … , A n }, having i = 1 , 2 , … , n alternatives, is evaluated by m decision criteria of set C j = { C 1 , C 2 , … , C m }, with j = 1 , 2 , … , m. Let w j = ( w 1 , w 2 , … , w m ) be the vector set used for defining the criteria weights, where w j > 00$]]> and ∑ j = 1 n w j = 1. Table 2 presents linguistic terms and their corresponding interval-valued fermatean fuzzy numbers (IVFFNs).

Step 2: Construct the pairwise comparison matrix Z = ( z i j ) m × m based on the opinions of experts given in Table 2. (33) Z = 1 z 12 ⋯ z 1 m z 21 1 ⋯ z 2 m ⋮ ⋮ ⋱ ⋮ z m 1 z m 2 ⋯ 1 , where z i j = [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] .

Step 3: Check for the consistency of each pairwise comparison matrix ( Z ). Here, to measure the consistency of expert judgments, match the crisp numbers obtained after defuzzifying to IVFFNs given in Table 2 based on Saaty’s scale. Then, apply the Saaty’s classical consistency process.

Step 4: Aggregate the judgments of experts.

The pairwise comparison matrix constituted for each expert is aggregated by using IVFFWG aggregation operator. Let E k = { E 1 , E 2 , … , E k }, with k = 1 , 2 , … , K, denote the set of experts having influence weights w k for each E k ; ∑ k = 1 K w k = 1. (34) IVPFWG ( z 1 , z 2 , … , z k ) = ( [ ∏ k = 1 K ( μ k L ) w k , ∏ k = 1 K ( μ k U ) w k ] , × [ ( 1 − ∏ k = 1 K ( 1 − ( υ k L ) 3 ) w k ) 3 , ( 1 − ∏ k = 1 K ( 1 − ( υ k U ) 3 ) w k ) 3 ] ) .

Step 5: Find the differences matrix D = ( d i j ) m × m between lower and upper points of the membership and non-membership functions using Eqs. (35) and (36): (35) d i j L = ( μ i j L ) 3 − ( ν i j U ) 3 , (36) d i j U = ( μ i j U ) 3 − ( ν i j L ) 3 .

Step 6: Find the interval multiplicative matrix S = ( s i j ) m × m Eqs. (37) and (38): (37) s i j L = 1000 d i j L 3 , (38) s i j U = 1000 d i j U 3 .

Step 7: Obtain the indeterminacy value T = ( t i j ) m × m of the z i j using Eq. (39): (39) t i j = 1 − ( μ i j U 3 − μ i j L 3 ) − ( ν i j U 3 − ν i j L 3 ) .

Step 8: Multiply the indeterminacy degrees with S = ( s i j ) m × m matrix to obtain the matrix of unnormalized weights R = ( r i j ) m × m using Eq. (40): (40) r i j = ( s i j L + s i j U 2 ) t i j .

Step 9. Obtain the normalized priority weights w i by using Eq. (41): (41) w i = ∑ j = 1 m r i j ∑ i = 1 m ∑ j = 1 m r i j .

Step 10. Rank the alternatives based on the normalized priority weights obtained in Step 9.

With the rapid advances in technology, digitalization has become an increasingly important issue investigated and discussed by academics and industries around the world. Currently, it has the potential to affect all sectors including supply chain and logistics. Thus, “digital supply chain” or “supply chain 4.0” has been introduced in the industrial world. Although many organizations have initiated a digital transformation in supply chains, they have not tackled it as a holistic approach to their DSC. This situation has caused delays in the progress of DSC until now. Hence, the biggest obstacle to successful digital transformation in the supply chain has been the lack of digital strategies in organizations. There has been a need to develop a framework and create awareness for the successful implementation of digital supply chain strategies. Due to this requirement emerging, organizations are required to evaluate their strategies according to certain criteria and they should use an MCDM method.

In this section, the digital transformation strategies in the supply chain are evaluated by utilizing the proposed method and it is aimed to choose the best strategy among various alternatives. A decision-making group of three experts is formed to evaluate the strategies using the proposed method. In a fuzzy environment, three decision makers, abbreviated as E1, E2 and E3, are selected, consisting of academicians who are experts in multi-criteria decision making. The weights of the decision makers are considered equal because they have the same level of experience. As a result of expert opinions and evaluation of the studies in the literature, three main criteria, fifteen sub-criteria and four alternatives have been determined for the strategies required for digital transformation in the supply chain. The determined main criteria are DC- Digital Competence, O- Organizational, and M- Management. The sub-criteria are listed as DC1- Digital Culture, DC2- Security and Privacy, DC3- Automation, DC4- Standardization, DC5- Innovation, O1- Sharing Information, O2- Cross-Functional Relationship, O3- Integration, O4- Training and Skills Development, M1- Adoption of Advanced Analytical Tools, M2- Supply Chain Visibility, M3- Financial Orientation, M4- Customer Orientation, M5-Flexibility, and M6- Enhanced Response. Alternative strategies are A1- Human Resource Management and Talent-Based Strategies, A2- Demand-Based Strategies, A3- New Business Models-Based Strategies, and A4- Technology and IT-Based Strategies. Fig. 5 illustrates this hierarchical structure involving the main criteria, sub-criteria, and alternatives. These alternatives and criteria are evaluated by constructing pairwise comparison matrices through linguistic terms given in Table 2 by three experts. The pairwise comparison matrices consisting of linguistic terms for the main criteria, sub-criteria, and alternatives are presented with the consistency ratio in Tables 3–21. The consistency ratios of the pairwise comparison matrices are calculated using the linguistic scale and corresponding numerical values given in Table 2. Due to space constraints, the next steps of the developed method are shown on the main criteria. After linguistic expressions in the pairwise comparison, matrices are converted to IVFFNs using the relevant scale, each expert’s assessment is aggregated with the IVFFWG operator. Table 22 presents the aggregated IVFF values of the main criteria. Then, IVFF-AHP is used to obtain the weights of criteria and alternatives. Table 23 gives the difference matrix D = ( d i j ) m × m between lower and upper values of the membership and non-membership degrees calculated based on Eqs. (35) and (36). The interval multiplicative matrix S = ( s i j ) m × m given in Table 24 is calculated based on Eqs. (37) and (38) in Step 6. The matrix of weights before normalization R = ( r i j ) m × m presented in Table 25 is obtained based on Eq. (40) in Step 8 by using the indeterminacy values given in Eq. (39). Then, the priority weights of each criterion obtained by using Eq. (41) in Step 9 and the final overall criteria weights are presented in Table 26. The overall criteria weights are obtained by multiplying the weights of the related main criteria and sub-criteria. Table 27 presents the priority weights of the alternatives according to the evaluation criteria. Finally, according to score values and ranking of alternatives demonstrated in Table 28, A2 is selected as the most suitable alternative. Demand-Based Strategies should be adapted with the largest priority, followed by New Business Models-Based Strategies, Technology and IT-Based Strategies, and Human Resource Management and Talent-Based Strategies.