Blockchain in healthcare supply chain

The traditional supply chain of the healthcare industry is being transformed through the use of a blockchain by ensuring data security and decentralization. Blockchain integration in the pharmaceutical supply chain leads to the elimination of counterfeit drugs as the system helps keep track of each product as it moves through the supply system from the manufacturer to the consumer [17]. This unalterable record provides openness, a key element in combating drug counterfeiting. Implementation of blockchain and smart contracts also helps to automate other operations, such as the verification of orders and payments, as well as increasing the level of security of health care activities [18]. However, there are number of limitations or challenges that has come in light while implementing the blockchain in healthcare supply chains. Including a range of stakeholders, a healthcare network is inherently large, and blockchain has scale limitations. The final issue is to overcome these infrastructural and technological barriers to achieve broader implementation [19]. This is especially true for pharmaceuticals where product identity is an important aspect due to potential dangers posed to human life [20]. Furthermore, blockchain has a positive impact on the electronic record of medical products and facilitates the tracking of the goods. Thus, through the implementation of blockchain, the healthcare providers make certain that the counterfeits are very limited since each transaction is recorded and authenticated. The increased transparency, in this case, provides an assurance that patients are administered authentic drugs [21]. The primary obstacles to blockchain implementation in healthcare are the volatility of the data security issue and the insufficient involvement of stakeholders. In promoting the usage of blockchain, such challenges need to be overcome [22]. Moreover, the application of blockchain with the present-day environments, like IoT and cloud-based structures, will present concerns for effective integration [23].

Fuzzy Delphi and Dombi LMAW Model

Fuzzy Delphi Method (FDM) is an advanced form of consensus building technique which augments traditional version of Delphi and considers a theory of Fuzzy sets so as to represent and measure the uncertainty expressed by experts and linguistic preferences [11]. FDM has been successfully applied in many fields such as education, behavioral science and technology management to identify, confirm, and rank factors in a systematic fashion using repeated expert surveys and fuzzy aggregation of answers [24]. The DOMBI model LMAW was introduced to evaluate Metaverse transformations in logistics using the Logarithmic Methodology of Additive Weights that integrates Dombi norms [25]. Spherical fuzzy sets are better suited for handling uncertainty and can therefore assist groups in making improved choices. The greater application of fuzziness in decision-making is because Dombi-t norm and t-conorm enhance aggregation in complex judgements [26]. In dealing with uncertain linguistic information space for improving accuracy and flexibility in decision-making, there has been an introduction of evolutionary algorithm based approach for constructing fuzzy Archimedean Dombi normed weighted operators [27]. During the decision making process, this technique enables an adequate capturing as well as ranking for criteria interdependences so that green suppliers can get selected with less mistakes [28].

Research gap

Blockchain-based solutions has been explored in infrastructure resilience and separately there are few studies that looks into the role of innovation in resilience [29,30]. In this regard, the study seeks to overcome this gap by creating and running a LMAW-based model focused on Dombi using the systematic approach of rank factors of blockchain adoption. In addition to providing decision makers with a solid tool for making decisions in the complex world of blockchain technology, this strategy goes beyond a theoretical understanding of the barriers and enablers of blockchain integration.

Most studies currently in existence employ traditional MCDM methods, such as the AHP or BWM, which, although reliable, frequently presume linear correlations and may oversimplify the subjective character of expert opinions, particularly in high-uncertainty settings like the healthcare industry in Bangladesh. This work pioneers the integration of the FDM with the Dombi LMAW in order to overcome this. The FDM achieves systematic expert consensus under fuzzy logic, thereby reducing the factor list to contextually relevant criteria. The Dombi operator provides more flexibility to model the non-linear interdependencies between factors and effectively accounts for the imprecision and vagueness inherent in linguistic expert evaluations, making the Dombi LMAW method superior for the subsequent prioritization. This ensures a more accurate and context-sensitive weighting system than traditional additive or multiplicative aggregation techniques. In complex fuzzy systems, this methodological combination offers a more reliable and theoretically valid tool for strategic decision-making.

Problem definition

This case study centers on Bangladesh’s healthcare system, which is essential to the country’s health. Blockchain technology has a lot of potential to improve healthcare supply chain management’s transparency, efficiency, and fraud reduction, but its implementation is being seriously impeded by resource limitations, legacy systems, and organizational opposition unique to poor nations. From a theoretical standpoint based on fuzzy systems, the diffusion factors that have been identified can be considered as crucial elements whose prioritization and selection are necessary to build a solid decision model for strategic success. Therefore, this study views the factor reduction process via FDM as analogous to heterogeneous feature selection under uncertainty [31], and the subsequent weighting via the Dombi LMAW model as an advanced mechanism for feature weighting and prioritization in a complex fuzzy environment. The fundamental problem addressed is thus the MCDM challenge under uncertainty: specifically, how to accurately weight and prioritize a large, competing set of diffusion factors (i.e., critical features) to determine the optimal, resource-efficient strategy for maximizing the blockchain diffusion effect in the Bangladeshi healthcare supply chain. This strategic prioritization is necessary to move from theoretical discussions of barriers to a clear, actionable policy roadmap.

Identification of blockchain diffusion factors

A comprehensive literature review was conducted to find out the most influential factors that could cause the increase in blockchain diffusion in the healthcare supply chain of Bangladesh. The theoretical background of blockchain diffusion was premised on past empirical research that has not only addressed the blockchain adoption in different fields (such as manufacturing, logistics, and healthcare) but also had a focus on blockchain adoption in the different fields. To a greater extent, the studies by [10,32,33] gave very important information on the contextual, technological and managerial complexities that blockchain implementation faces in emerging economies. Upon the foundation of these and some other relevant research, the existing study has determined five core categories of factors that enhance blockchain diffusion considerably: Organizational Factors (OF), Technological Factors (TF), Environmental Factors (EF), Managerial Factors (MF), and Individual Factors (IF). The final list of contextually relevant factors, along with their detailed descriptions and sources, are consolidated in the streamlined Table 3.

Organizational factors (OF)

This dimension, encompassing internal capacities and preparedness, covers the following factors: Organizational Readiness (OF1), Implementation Cost (OF2), Resistance to Change (OF3), Top Management Support (OF4), Lack of ROI Clarity (OF5) and Lack of Practical Use Cases (OF6) [33–35].

Technological factors (TF)

This dimension deals with the blockchain system’s fundamental technical features and compatibility problems. These factors encompass: Blockchain Scalability (TF1), Interoperability with Legacy Systems (TF2), Smart Contract Automation (TF3), Integration with IT Infrastructure (TF4), Standardization Issues (TF5), Traceability (TF6) and Cybersecurity Risks (TF7) [10,34–38].

Environmental factors (EF)

External pressures and regulatory circumstances are related to environmental factors. The identified factors are: Competitive Pressure (EF1), Innovation Pressure (EF2), Vendor Readiness (EF3), Regulatory and Legal Clarity (EF4) and Performance Expectancy (EF5) [34,35,39].

Managerial factors (MF)

The Managerial Factors focusing on strategic leadership and internal governance, consist of: Transparency and Immutability (MF1), Perceived Benefits (MF2), Trust Among Stakeholders (MF3) and Disintermediation (MF4) [10,40].

Individual factors (IF)

User-centric and behavioral elements are captured by the following factors: Lack of Awareness (IF1), Trust in Technology (IF2), Social Influence (IF3) and Effort Expectancy (IF4) [33,41].

Research methodology

A two-phase methodology is used in the study framework to determine and rank the major elements impacting the adoption of blockchain technology in healthcare supply chains. A thorough assessment of the literature and expert consultations are used to first identify important factors. Fuzzy-based scoring is then used to systematically rank these aspects, allowing for a thorough assessment of their relative importance. In order to ensure dependability in the prioritization process, sensitivity analysis is finally performed to assess the stability and robustness of the outcomes. This approach makes it easier to comprehend in detail the key factors influencing blockchain adoption in the healthcare industry. The two-stage research framework is depicted in Fig 1.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. The methodological framework.

https://doi.org/10.1371/journal.pone.0348578.g001

Participants and study design

An observational and cross-sectional analysis were used to conduct the investigation. A panel of ten experts from four distinct healthcare organizations was chosen using a purposive and judgmental sampling technique. These experts were divided into five categories for analysis. This non-probabilistic method was used to make sure that each participant had in-depth knowledge of a particular field that was pertinent to the study questions. The ten participants’ familiarity with the issues that arise in the operation of the healthcare industry adds to the data’s trustworthiness. To mitigate potential biases, the expert pool was intentionally diversified to cover different organizational types (Government, Private, Academic and Technical) and functional roles, including Clinical (Senior/Junior Doctor, Nurse), Managerial (Hospital Manager, Supply Chain Manager), Technical (IT Specialist, Pharmacist) and Policy (Govt. Health Official, Academic Researcher). Instead of utilizing statistical significance testing, which is typically irrelevant for consensus-building procedures using judgmental sampling in a qualitative-dominant framework, the sample size of 10 experts was chosen based on the expert-driven consensus principle inherent in the Delphi method. In such methodological frameworks, the validity of the results depends on the specialized knowledge of the participants rather than the sample size; as noted by [42], a panel of 10–15 experts is generally sufficient to yield stable results and reach a meaningful consensus in a Delphi-based study. Furthermore, this size was sufficient to cover all four critical functional roles (Clinical, Managerial, Technical, and Policy) across different organizational types (Government, Private, Academic) essential for ensuring construct validity and mitigating bias. Anonymity was maintained and all participants provided written informed consent after receiving a study briefing. Table 1 summarizes participant profiles and their relational assessments. This study involved the prospective recruitment of an expert panel to provide professional opinions via anonymous questionnaires regarding organizational and technical factors influencing blockchain diffusion in healthcare. The study posed minimal risk, was not a patient-based study, and did not collect any personally identifiable participant information or sensitive health data. Therefore, ethical approval was determined to be unnecessary by the Office of the Director (Research & Extension), Khulna University of Engineering & Technology, Khulna-9203, Bangladesh. The expert recruitment period spanned from April 1, 2024 to July 30, 2024. All participants (aged 18 and up) provided informed written consent electronically, confirming their understanding of the study’s purpose and confidentiality protocols. Anonymity of all responses was maintained throughout the data collection and analysis.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Overview of the experts’ profiles.

https://doi.org/10.1371/journal.pone.0348578.t001

Data collection procedure

The data collection was executed in two distinct stages aligned with the FDM and Dombi LMAW model. Participants were initially contacted and onboarded by the authors via telephone and email. The primary data collection instrument was a structured Google Form questionnaire, which was electronically distributed to all 10 experts. This form contained comprehensive instructions regarding the linguistic scales and the specific manner in which the criteria should be evaluated. All participants completed the questionnaire on the MCDM techniques. The comprehensive list of the initial 26 Blockchain Diffusion Factors identified for expert evaluation in the Fuzzy Delphi Method is shown in S1 Table in S1 File. The estimated time required for each participant to complete the questionnaires was approximately forty to forty-five minutes. The entire data gathering process took an approximate of four months to accomplish.

Data validity and questionnaire reliability test

To construct the determination of the relevance of the blockchain diffusion factors in the healthcare supply chain, the fuzzy Delphi Method was used to obtain 10 expert responses. One of them represented the pharmaceutical field, two were connected with hospital supply chain and logistics management, three had experience in healthcare operations and technology (including clinical workflows, patient care, and coordination), two were affiliated with an academic or research institution with a focus on blockchain or analytics, one was an IT Healthcare operations manager, and one was a government health expert experienced with public health supply and policy. The survey included 30 questions; 26 questions were associated with the given factors and 4 questions were the demographic questions designed to describe the professional background and level of experiences of the respondent. Cronbach alpha was used to predict the internal consistency and reliability of the instrument and the measure was 0.812, which is better than the set standard of 0.7 and shows good level of reliability [33]. The data analysis was done by processing it using IBM SPSS statistics (Version 22.0). The non response bias was to be examined by doing chi-square analysis with the responses being categorized as early submissions and late submissions coded according to the time that they were completed. Statistically, the outcomes displayed greater similarities between the two groups, and thus a contribution of nonresponse bias was not evident, since the data specified by the collected expert opinions could be substantiated efficiently [43].

Fuzzy Delphi Method

The Fuzzy Delphi Method (FDM) is the modified variant of Delphi that considers the linguistic preferences of people regarding decision making. This it done by combining fuzzy theory with the conventional Delphi. It takes the opinions of the experts and uses linguistic variables each of which is one of the following: very high, high, moderate, low and very low in a number of rounds of anonymous questionnaires and answers. This method then employs fuzzy set procedures to aggregate such responses after these language variables are converted into fuzzy sets using fuzzy membership functions [44]. The fuzzy Delphi approach is explained briefly in the following:

Establishing the linguistic variables’ triangular fuzzy number

A fuzzy set with a triangle membership function is called a triangular fuzzy set. The following is the equation for a triangular fuzzy set:

(1)

A triangular fuzzy number is defined as in which i ∈ {1, 2, … n} denotes the experts and j ∈ {1, 2, … m} denotes the factors, where n and m are the number of experts and factors respectively. The fuzzy number set of the linguistic variables is obtained from the research framework presented in Table 2 [45].

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Establishing the Fuzzy Number Set for Linguistic Variables.

https://doi.org/10.1371/journal.pone.0348578.t002

Collecting experts’ opinions

After the identification of the factors, a questionnaire containing the linguistic factors provided in the above table is administered to four experts from academia and industry to measure the importance of the factors. Fuzzy triangular numbers in this study are applied in measuring the various factors. The method also utilizes a geometric mean method to determine the group choice by the experts [46].

Defuzzification

The procedure for converting a fuzzy set into a suitable crisp set is called defuzzification. This method is often used when using fuzzy logic controllers. It involves selection of a defuzzification method depending on the problem-solving system of the actual problem being solved, mean-max, centroid, maximum, or weighted average. This procedure is required in order to convert the rather general control signals into definite values [47].

The geometric mean aggregation of “n” expert opinions results in a single aggregated TFN for each factor j. The components are calculated using the geometric mean of the lower, median, and upper bounds of the individual TFNs provided by the experts:

(2)

Where

In this study, the expert group choice is ascertained through the application of a geometric mean model. The fuzzy weights are defuzzified using the simple center of gravity approach to yield a crisp value (CV_j), which may be calculated using the following formula:

(3)

i ∈ {1, 2, …., n} denotes the experts and j ∈ {1, 2, …., m} denotes the factors.

Determination of significant factors

The final step in the fuzzy Delphi technique is to identify the substantial factors, which is done by comparing the weight of each factor to the “S” threshold. The value of is calculated by averaging the weight of each factor. The screening principle is as follows:

(4)

Factor is chosen if ^CV_j ≥ S.

Factor is rejected if ^CV_j < S.

To be compared, the values need to be transformed into crisp values. Following the defuzzification process, the consensus threshold, S was calculated as the average of the crisp values (CV_j) for all 26 initial factors. The determined threshold value used for screening was S = 6.9, as calculated in S3 Table in S1 File. The screening principle was then applied: a factor was retained for the second stage of the study if its crisp value met or exceeded the threshold (CV_j ≥ S), indicating strong expert consensus on its significance. Conversely, any factor with a score below the threshold (CV_j < S) was eliminated as being contextually less relevant to the Bangladeshi healthcare sector. Applying this logic reduced the initial list from 26 factors to 22 significant factors. This filtering process ensures that the subsequent Dombi LMAW prioritization focuses exclusively on high-impact variables, thereby enhancing the analytical precision of the model. Fig 2 illustrates the Fuzzy Delphi Methodology for achieving consensus in healthcare supply chain concerns.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Methodology of Fuzzy Delphi method.

https://doi.org/10.1371/journal.pone.0348578.g002

Dombi Operator based LMAW

This section provided the preliminaries and basic operations of Dombi T-norms/T-conorms under triangular fuzzy numbers (TRNs).

Dombi T-norm and T-conorm

In this study, triangular fuzzy numbers are applied to address the uncertainty in the information [48,49].

The notions and operations of the Dombi T-norm and T-conorm were introduced which present the advantage of good flexibility with the operational parameter [50]. Some definitions of the Dombi T-norm and T-conorm under triangular fuzzy numbers (TFNs) are expressed by:

Definition 1. Let and be any two real numbers. Then, the operations of Dombi T-norm and T-conorm between and are expressed by:

(5)

(6)

Where Υ > 0 and (, ) ∈ [0,1].

Definition 2. Let , and be two TFNs, Υ > 0 and let it be a fuzzy function, then some operational laws of TFNs based on the Dombi T-norm and T-conorm are defined by:

1) Addition of two fuzzy numbers and can be defined as follows:

(7)

2) Multiplication of two fuzzy numbers and can be defined as follows:

(8)

3) Scalar Multiplication,

(9)

4) Power,

(10)

κ>0

Definition 3. Let , a set of TFNs, and ∈ [0, 1] denotes the weight of coefficients of , which fulfills the requirement that it is . Then, the fuzzy weighted averaging (FWA) operator and fuzzy weighted geometric averaging (FWGA) operator are expressed by:

(11)

(12)

Determining Factors weights — fuzzy Dombi-based LMAW

The Dombi based LMAW methods are used to determine the weights of factors. The proposed Dombi based LMAW follows two consecutive stages.

(1) Framework definition

Step 1. The objectives of the decision-making criteria are explained, along with the factors that will be chosen and the experts that will be assembled to construct the suggested model. The Factors = (, , …, ) (j = 1, 2, …, m) are evaluated by a set of experts = (, , …, ) (l = 1,2, …, e).

Step 2. The linguistic terms and their corresponding values are determined.

(2) Determining Factors weights – DOMBI LMAW

Step 3. The steps of this model are summarized as follows:

Step 3.1. Firstly, the fuzzy priority vector is described. The factors are evaluated by the experts with the help of fuzzy linguistic terms and their corresponding values (see Table 1). Then, the fuzzy priority vector is obtained, where (1 ≤ ℓ ≤ e).

Step 3.2. The absolute anti-ideal point () is determined.

(13)

The relationship between the fuzzy priority vector elements and the is found by:

(14)

Where ( represents the elements of the priority vector η.

Step 3.3 The vectors of weight coefficients are found by Eq. (15):

(15)

Where

Step 3.4 The aggregated fuzzy priority vector is defined by:

(16)

Where σ > 0, and

Step 3.5 The triangular fuzzy value of is defuzzified by:

(17)

The used model is proposed by [51]. The flow chart of the used model is shown in Fig 3.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Dombi LMAW method.

https://doi.org/10.1371/journal.pone.0348578.g003

Case study

With an emphasis on strengthening resilience in the face of adversities like pandemics, this study sought to investigate the elements supporting the adoption of Blockchain technology within the supply chain management of the healthcare sector in Bangladesh. To evaluate the several aspects impacting the diffusion of Blockchain technology in healthcare supply chain management, firstly Fuzzy Delphi Method was utilized to identify the substantial factors, where the assessments of the factors are reported in S2 Table in S1 File and the defuzzified scores of the factors are provided in S3 Table in S1 File. Then a fuzzy Dombi-based decision-making model was utilized. In order to increase resilience, this case study focuses on the healthcare sector in Bangladesh and attempts to pinpoint the elements that make a blockchain implementation successful. Total four healthcare were examined for this scenario. The initial set of 26 factors was reduced to 22 significant factors through the Fuzzy Delphi screening process. The consensus threshold, S was calculated as the average of the defuzzified crisp values (CV_j) of all 26 initial factors, yielding S = 6.9. The four factors eliminated from the analysis, having been assessed by the expert panel as contextually less relevant (i.e., their CV_j < S), were: Lack of Practical Use Cases (OF6), Cybersecurity Risks (TF7), Performance Expectancy (EF5) and Effort Expectancy (IF4). The exclusion of these factors suggests that in the current resource-constrained healthcare landscape of Bangladesh, experts prioritize foundational organizational readiness and leadership support over technical secondary risks or individual behavioral expectations. This systematic reduction ensures that the subsequent Dombi LMAW prioritization focuses solely on the factors deemed most critical for the Bangladeshi healthcare context. The finalized list of blockchain diffusion factors obtained from Fuzzy Delphi method is shown in Table 3.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Finalized list of blockchain diffusion factors in healthcare sector.

https://doi.org/10.1371/journal.pone.0348578.t003

Healthcare A

Healthcare A is a government-owned emergency ward hospital in Khulna, picked because it represents the scenario of the government-based healthcare activities in a big-city setup. The data gathered in this hospital was to get to know the issues associated with the state-run supply chain digitalization. The hospital had challenges because although it was staffed by experienced professionals, it was having challenges of old data systems, bad interoperability and poor organizational readiness. Using the Fuzzy Delphi and Dombi LMAW methodology, possible weaknesses were outlined (especially in stakeholder trust and infrastructure preparedness) and could be useful in the adoption of blockchain in a similar public institution by offering opportunities where action can be taken.

A questionnaire is conducted to evaluate the Factors. Each criterion is scored by the experts. The questionnaires are sent to ten experts. The results are then gathered. Twenty-two Factors = (1, 2,..., 22) are evaluated by a set of experts = (1, 2,....,10) with the help of Table 4 which presents the linguistic scale to obtain the experts’ opinions [52].

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Fuzzy linguistic scale.

https://doi.org/10.1371/journal.pone.0348578.t004

For Healthcare A, the experimental results are shown below:

Step 1. The fuzzy priority vectors are defined by ten experts in this step. The obtained values are given in Table 5.

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Fuzzy Priority Vectors for Healthcare A.

https://doi.org/10.1371/journal.pone.0348578.t005

Step 2.

Using Eq. (13) and expressing the relationship vector, the value of AIP= (0.4, 0.5, 0.6) is integrated.

Later, AIP is applied to obtain the ratio vector by Eq. (14). The results are given in Table 6. For example, the ratio vector for OF1 is calculated by Eq. (14) as follows:

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Factors ratio vectors for Healthcare A.

https://doi.org/10.1371/journal.pone.0348578.t006

The data of the Factors Ratio Vectors for Healthcare A from the remaining five experts (E6-E10) is shown in S4 Table in S1 File.

Step 3.

The final vectors of weighting coefficients are calculated in this step.

For example, the weighting coefficients for OF1 is calculated by Eq. (15) as follows:

Step 4.

The aggregated fuzzy priority vector is calculated by Eq. (16). For example, the aggregated fuzzy priority vector for OF1 is,

Where, and σ

Step 5.

The Scores are obtained by defuzzifying the aggregated local fuzzy weights by Eq. (17).

For example, the score of OF1 is,

Then, the ranks of the factors are determined from 1 to 22 through the descending order of the scores. The complete results of the prioritization for Healthcare A are shown in Table 7.

Download:

• PNG
  larger image
• TIFF
  original image

Table 7. The local fuzzy values, scores and ranks of the factors for Healthcare A.

https://doi.org/10.1371/journal.pone.0348578.t007

Fig 4 shows the final scores of Blockchain Success factors for diffusing in Healthcare A supply chain.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Final scores of Blockchain success factors in Healthcare A.

https://doi.org/10.1371/journal.pone.0348578.g004

Healthcare B

Healthcare B is located in Rajshahi and has been selected as a second example of the representative treating institution by the geographic area to gain comparative patterns about the regional differences in digital readiness. Regardless of the fame and magnitude, the hospital experienced a disjointed communication flow, unstandardized processes of supplies, and financial restrictions. To the extent that the inclusion of this site served to confirm that such heavyweight factors of diffusion as management support and regulatory clarity were indeed found to have a consistent importance across regions. The outcome of the study is bound to shape directional measures that will engage these common public sector challenges. Table 8 shows the final experimental results for Healthcare B.

Download:

• PNG
  larger image
• TIFF
  original image

Table 8. The fuzzy scores and ranks of the factors for Healthcare B.

https://doi.org/10.1371/journal.pone.0348578.t008

In the prioritization of the factors across the healthcare organizations, Healthcare B has experienced mild changes in ranks relative to Healthcare A: there is interchange of ranks between TF5 (Standardization Issues) and IF1 (Lack of awareness) where TF5 ranks 18^th and IF1 ranks 17^th in Healthcare B. There is also interchange between MF1 (Transparency and immutability) and IF2 (Trust in technology) in which MF1 is 11^th and IF2 is 10^th. Such shifts signify regional and contextual differences in the way technological, managerial and individual aspects are understood and valued within the healthcare supply chain.

Healthcare C

Healthcare C is a leading middle-sized privately owned hospital in Khulna. It was chosen to give an understanding of how moderately scaled independent institutions treat the assimilation of the emerging technology. Although this facility has relative flexibility in its operations, it has also been facing challenges like mistrust and resistance to change by the stakeholders. Data gathered in this hospital enhanced the analysis by showing how even in well-equipped surroundings non-technical barriers are still present. The results of the prioritization advised the hospital to initiate regularized awareness and training to smoothen the road to blockchain adoption. Table 9 shows the final experimental results for Healthcare C.

Download:

• PNG
  larger image
• TIFF
  original image

Table 9. The fuzzy scores and ranks of the factors for Healthcare C.

https://doi.org/10.1371/journal.pone.0348578.t009

In Healthcare C, the placement of TF1 (Blockchain Scalability) and TF2 (Interoperability with Legacy Systems) have swapped meaning that the priority has shifted in the perceived importance of the technical infrastructure concerns. Likewise, EF4 (Regulatory and Legal Clarity) and IF3 (Social Influence), have also exchanged places which implies that stakeholders’ social dynamics are regarded to be a bit more relevant than regulatory clarity as far as blockchain uptake is concerned.

Healthcare D

Healthcare D is one of the high-end, exclusive medical service providers, located in Khulna as well. It was added to understand how even technologically advanced organizations struggle with the ability to ensure that blockchain investments support the perceived revenue and internal capacity. Including this hospital in the decision enabled the research to grasp the fact that digital maturity is not always characterized by blockchain adoption readiness. The knowledge acquired served to reinforce the value of leadership commitment and understanding the return on investment, especially where the business viability is at stake, like in the private sector. The final experimental result for Healthcare D is shown in Table 10.

Download:

• PNG
  larger image
• TIFF
  original image

Table 10. The fuzzy scores and ranks of the factors for Healthcare D.

https://doi.org/10.1371/journal.pone.0348578.t010

In Healthcare D, only a significant change in factor ranking is the swap between MF1 (Transparency and Immutability) and OF2 (Implementation Cost), where MF1 moves up to 8^th position and OF2–10^th. Such transition indicates that in such an auxiliary high level of privatization in the private healthcare industry, the value of transparency and immutability of blockchain properties overshadows higher-order cost efficiency, another indication of putting long-term working integrity before a more short-term cost-driven paradigm. The rest of the factors retain their former rankings.

Sensitivity Analysis

In this section, the analysis of the parameter based on the subjective evaluations of the experts in the decision-making problem was made. 50 experiments were carried out for in fuzzy Dombi parameter to calculate the factors weights. It was observed if the ranking of the factors were changed in changing the value of that is from the interval 1 ≤  ≤ 50. Table 11 represents the final score and ranking of the Healthcare A with new value of , which is 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 11. Rank of the factors for σ = 2 for Healthcare A.

https://doi.org/10.1371/journal.pone.0348578.t011

Fig 5 illustrates the change in the value of , and the new scores and ranking of the factors for the Healthcare A.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Influence of σ on the ranking of factors for Healthcare A.

It can be seen that there was no change in the rank of the factors.

https://doi.org/10.1371/journal.pone.0348578.g005

Theoretical Implications

This study’s prioritization of Top Management Support (OF4) and Organizational Readiness (OF1) as the top two factors carries significant theoretical implications. While our MCDM approach determines the relative weight or importance of these factors, their consistent top ranking aligns with established diffusion and technology adoption theories, such as the Technology-Organization-Environment (TOE) framework [36]. According to these models, organizational elements like leadership support and preparedness serve as the cornerstone of any adoption process [35].

Additionally, the results empirically support the Organization component of the TOE framework as the main area of strategic focus in emerging economies, surpassing pressing issues with the Environmental (e.g., Innovation Pressure) and Technology (e.g., Smart Contract Automation) dimensions. This study makes a contribution to fuzzy set theory by showing how the Dombi operator produces a prioritization hierarchy that is more stable and theoretically justified than those generated by models assuming linear compensation among criteria by taking into account the high vagueness and non-linearity inherent in expert judgment. These elements serve as essential, non-compensatory threshold conditions for the successful diffusion of blockchain, especially in emerging economies [33]. In other words, a significant lack of leadership support or fundamental readiness cannot be made up for by a high score in other factors (such as Innovation Pressure or Smart Contract Automation). While this study did not execute a formal Necessary Condition Analysis (NCA) to statistically test this causal attribute; the findings strongly suggest that OF4 and OF1 are necessary conditions. Future research should aim to formally test this hypothesis using NCA to enhance the theoretical persuasiveness of the prioritization.

Managerial Implications

Blockchain technology represents an excellent opportunity for textile and healthcare manufacturers to advance their supply chains in terms of transparency and traceability. Manufacturers will get blockchain systems that will log every movement of raw materials, components, and finished goods, providing a clear visibility of the overall production process. Such transparency helps manufacturers track inefficiencies, make counterfeit goods less common and contribute to the best practices of supply chain management. Blockchain technology will help to streamline differences between healthcare suppliers and manufacturers in the procurement processes and improve collaboration among them. Instead of using the blockchain-based smart contracts to enable interactions with purchase agreements, payments and delivery schedules, the number of parties taking administrative tasks and disputes can be significantly reduced. Furthermore, financing solutions of blockchain-based supply chains can be offered to suppliers on affordable credit basis, thus making it easy for them to surmount the cash flow bottlenecks and post sustainable business growth [54]. Retailers within the healthcare sector can form trust and loyalty of customers by implementing blockchain technology in their services. Through enabling customers to see clear and verified information concerning the providence of items, the constituents and the manufacturing process, retailers are able to separate their brands and be confident to attract morally oriented customers. Also, transactions that are made via blockchain-based loyalty programs or rewards systems can encourage customers to make repeat purchases and hence create more bonds between the organizations and their customers [55]. Policymakers and industry body representatives are among the major drivers in the healthcare sector’s blockchain adoption. Through the advancement of such industry-wide standards and guidelines for blockchain implementation, policymakers are able to ensure a refined legal framework that allows the promotion of new innovations and opportunities in collaboration. In addition, lawmakers should also explore how to use blockchain-based solutions for legal compliance, tax collection, and worker’s rights protection, which will improve the sector’s sustainability and integrity [56]. The blockchain technology in healthcare sector can bring not only revolutionary effects but also help increase supply chain transparency and leveraging the consumer trust. Integration of blockchain innovation is a key to value creation, strengthening of competitive positions and sustainable value for different revolves of industry.

Limitations and future scopes

The study is based on relatively small sample of the expert opinions that can limit the applicability of the conclusions to other settings. Besides, even though 26 factors were identified within the context of the literature, there are other factors that might have been significant and which were not included because of contextual narrowing. Specifically, factors such as Vendor Support, Data Governance Frameworks and Digital Literacy were determined to be less immediate priorities by the expert consensus during the initial screening, reflecting the current low digital maturity level in the Bangladeshi healthcare sector where foundational organizational (OF) and managerial (MF) elements take precedence over advanced ecosystem and regulatory constructs [33,35,36]. This study strongly advocates for including these factors in future cross-country research to compare how blockchain diffusion priorities shift across varying national digital maturity levels.

Future work can also apply dynamic modeling methods (e.g., Fuzzy DANP or System Dynamics) to rigorously explore the evolving, time-dependent interdependencies among factors. Furthermore, future research should integrate parallel MCDM methods (such as AHP, BWM, or TOPSIS) to empirically compare rank stability and validate the superiority of the Dombi LMAW weighting in complex fuzzy environments, thereby enhancing the methodological discourse. The economic development level can be quantified using metrics such as Gross Domestic Product (GDP) per capita or the World Bank’s income-group classification and the medical informatization level can be quantified using proxies such as national healthcare IT spending (as a percentage of GDP), the Digital Adoption Index (DAI) or EHR (Electronic Health Record) adoption rates within the country.