Blockchain Development Services Provider Assessment Model for a Logistics Organizations

Abstract

As the global market develops along with technological advances, especially the opening of markets in developing and underdeveloped countries, the logistics industry is considered by managers to be a useful tool to link different functions of corporate strategy. Logistics is the process of planning, implementing and controlling the movement of goods or information that are related to raw materials (inputs) and final products (outputs) from the point of origin to the point of consumption in order to meet customer requirements. The logistics industry brings about many positive effects, not only for the national economy but also for creating equal business opportunities for businesses of all economic sectors. However, large companies in the logistics industry still depend on EDI or APIS to exchange secure authentication data, which increase the security of operations within the industry, but this is the cause of a lot of confusion with serious consequences on the supply chain when it is too dependent on these systems. To minimize the risks and optimize the problems mentioned above, blockchain technology could help the logistics industry to operate optimally. In an era when everything is digitized, from personal information to financial transactions, choosing a system with high reliability and security becomes extremely important. Currently, there are many blockchain development services providers and it is essential to choose the investment in this technology that suits the characteristics and purposes of the logistics organization in question. This study aimed to propose a two-stage fuzzy multicriteria decision-making model for the assessment of blockchain development services providers for logistics organizations. The hybrid MCDM model was evaluated through a real-world case study at a logistics organization. The contribution of this work is the provision of useful guidelines for the evaluation and selection of blockchain technology services providers for logistics companies.

1. Introduction

Over the past 20 years, Vietnam has become one of the most popular manufacturing hubs in Asia thanks to its steady growth rate, the import–export-oriented economy, increasing FTAs, a young dynamic workforce, investment incentives and strategic location. The recent US–China trade war, numerous free trade agreements and the relocation of multinational companies out of China in 2020 have moved Vietnam up the value chain [1]. However, the economy can only develop synchronously and smoothly when the logistics chain operates continuously. Therefore, the role of logistics within the economy is being increasingly promoted. Logistics has become the driving force for the flow of economic transactions and is also an important activity for most goods and services.

Logistics is the process of planning and executing the efficient transportation and storage of goods from the point of origin to the point of consumption. The goal of logistics is to meet customer requirements in a timely and cost-effective manner. Currently, most leading companies within the logistics industry are still dependent on EDI or APIS. These tools are used to exchange secure authentication data, which increases the security of operations in the industry. However, this mode of operation is the source of many errors, which can have serious consequences on the supply chain when it is too dependent on these systems. Some typical errors in logistics organizations can also be encountered, such as: errors in information when entering data; problems with linking systems, synchronizing data or controlling multiple repositories; and issues related to the management and control of the delivery time and quality of goods that have not been optimized [2]. Although the 4.0 revolution has contributed to the creation of ships with huge capacities and unprecedented speeds, seaports are automated by robots and huge databases that track the journey of goods, technology and services. However, the paperwork, which is an essential part of global trade, still seems to be slow and creates too much of a burden for businesses. Faced with this situation, blockchain technology is a potential solution to the problem of operational efficiency, as well as contradictions within the logistics system, especially in the current situation with the disruption of the supply chain of goods due to the COVID-19 pandemic. These things put managers in the new position of needing to solve the difficulties that the supply system is facing.

Blockchain technology is an ever-growing, secure, shared record-keeping system in which each user of the data holds a copy of the records, which can only be updated when all parties that are involved in a transaction agree to update them. The blockchain technology workflow creates a digital trail of all transactions within the organization, thereby making transactions almost impossible to falsify and audits much smoother. The blockchain process is shown in Figure 1 [3].

With its smart features, the use of blockchain technology in logistics helps to speed up the transportation process and saves a lot of time and money thanks to the automatic operation process and management, which help to manage the transportation process without any errors and effectively manage product status, such as temperature, usage time, condition, etc. Blockchain technology also contributes to limiting the incurred costs, such as wharves.

The complexity of the criteria for choosing a suitable blockchain services provider turns the problem into a multicriteria decision-making problem. In reality, the methods for solving MCDM problems are completed in two phases: the allocation of weighting ranks for the criteria being considered and an overall ranking of all of the alternatives compared to the calculated weights of the considered criteria. These methods are commonly carried out during supply chain strategy planning in order to solve conflicting objectives that involve the need for compromise. The author proposes a two-stage fuzzy multicriteria decision-making model, which includes the fuzzy analytic hierarchy process (FAHP) and the weighted aggregated sum product assessment model (WASPAS), for the assessment of blockchain development services providers for logistics organizations. The proposed model calculates the weight of each criterion based on the analytic hierarchy process (AHP) and then inputs these weights into the WASPAS model to rank the various blockchain logistics services providers.

The rest of this paper is structured as follows. Section 2 describes the relevant literature about the applications of MCDM models in many scientific fields. Section 3 presents the basic theory of two MCDM models. In Section 4, the proposed model is utilized for the case study of a real-world logistics services provider evaluation to demonstrate its feasibility. Section 5 concludes the research.

2. Literature Review

The supply chain is currently shifting toward the digital era, with continuous technological improvements digitalizing every aspect. With the growth of areas such as e-commerce, shipment tracking and many others, blockchain technology is also increasing in popularity due to its ability to provide supply chain leverage in this time period. However, there is still only a limited number of studies that have been conducted on the blockchain supplier selection problem of optimizing the best decision for businesses.

A study conducted by Özkan et al. investigated the risks of blockchain technology from multiple perspectives in order to rank those risks using the Pythagorean fuzzy analytical hierarchy process (PF-AHP) combined with the Delphi method, which consults experts in the field. The study showed that this method was able to rank critical risks successfully [4]. Another study in the blockchain industry that was conducted by Zafar et al. examined the immaturity of the blockchain field and proposed a number of combined multicriteria decision-making (MCDM) models, particularly the ECWM, which combines the entropy and CRITIC methods to rank the most suitable blockchain technology platforms [5]. Kaska and Tolga looked at blockchain software selection for the maritime industry and applied a basic ranking MCDM technique, which is popularly known as TOPSIS, to determine the most suitable software option amongst the wide range of available software [6].

There are also other studies that have discussed the different MCDM techniques that have been applied in the blockchain technology industry. Murat et al. utilized a hesitant fuzzy set incorporated with AHP and TOPSIS to determine a performance evaluation of the criteria for selecting the most suitable blockchain technology and compare suitable alternatives [7]. Karaşan et al. also used a similar method to evaluate the risks of blockchain technology using hesitant Z-fuzzy linguistic terms to assess the risks accordingly [8]. Deepu, on the other hand, used the IOIS method to look at how information systems could help to assess decision-making in the digitalization of the supply chain using blockchain technology [9].

Other research has also been conducted to show the current weaknesses in blockchain technology evaluation and supply chain digitalization fields for decision-makers [10,11,12,13,14,15]. This research aimed to provide a thorough insight into other MCDM models that could be used for the selection of a suitable blockchain technology alternative. Barbarosoglu et al. [16] used the AHP model to solve the supplier selection problem for a motor manufacturing company. Murugesan et al. [17] developed a new composite model using structural equation modeling (SEM) and AHP for the selection of suppliers. Based on the output from the composite model, a cluster analysis was then carried out to find the strengths and weakness of each supplier in terms of the influencing factors.

Ilieva et al. [18] proposed a conceptual framework for the group multicriteria selection of blockchain software in fuzzy environments according to organizational needs and expert judgements. Lai et al. [19] introduced a fuzzy MCDM model that included the integration of linguistic D numbers (LDNs), the double normalization-based multiple aggregation (DNMA) method and the criteria importance through intercriteria correlation (CRITIC) method for blockchain platform evaluation. Lin et al. [20] proposed an MCDM model that included AHP, TOPSIS and linear programing for an enterprise resource planning (ERP) system.

From this literature review, it was concluded that an MCDM would be the optimal technique for applications in complex situations that include multiple criteria and conflicting goals. This tool has received attention in all industries because of its flexibility for decision-makers with multiple problems. While there have been some studies on the application of MCDM techniques in solving problems in logistics organizations, few of them have considered the use of the MCDM model, especially in fuzzy decision-making environments. Thus, in this study, a fuzzy MCDM model was proposed for the assessment of blockchain development services providers for logistics organizations.

3. Methodology

3.1. Development of Research

The model development process included three stages, the research graph is shown in Figure 2.

Step 1: Examine and assess the current procedures (blockchain logistics services providers) using scientific studies and industry expertise (IT managers, blockchain technology experts, logistics operation managers, etc.) to gather more criteria for each challenge;

Step 2: For each challenge, create an MCDM model.

Generate the defined criteria weights using a fuzzy AHP for the blockchain logistics services provider selection problem and then calculate the ranking of the probable solutions using WASPAS;

Step 3: Discuss real-world case studies.

3.2. Theory foundation

3.2.1. The Fuzzy Analytical Hierarchy Process

A triangular fuzzy number (TFN) can be defined as $\left(n, p, q\right),\mathrm{where} n, p$ and $q \left(n \le p \le q\right)$ are the parameters that specify the smallest possible value, the most promising value and the largest possible value of the TFN, respectively. A triangular fuzzy number (TFN) can also be defined as $\left(f, h, g\right), \mathrm{where} f, h$ and $g \left(f \le h \le g\right)$ are the parameters that determine the smallest possible value, the most promising value and the largest possible value of the TFN. Figure 3 illustrates a typical TFN.

$$\mu \left(\frac{x}{\tilde{M}}\right)=\{\begin{array}{c}0,\\ \frac{x-f}{h-g}\\ \frac{g-x}{g-h}\\ 0,\end{array} \begin{array}{c}x<h,\\ f\le x\le h,\\ h\le x\le g,\\ x>g,\end{array}$$

(1)

A hazy number is defined as:

$$\tilde{M}=(M^{t\left(y\right)},M^{u\left(y\right)})=\left[f+\left(h-g\right)y, g+\left(h-g\right)y\right],y \in \left[0,1\right]$$

(2)

The fuzzy analytical hierarchy process (FAHP) is the fuzzy extension of the AHP to handle its limitations when working with uncertain decision-making environments. Let $X=\left\{x_1, x_2,\dots .x_n\right\}$ be the object set and $K=\left\{k_1, k_2,\dots .k_n\right\}$ be the goal set. According to the extent analysis method that was proposed by Chang [21], each object is considered and an extent analysis of its goals is performed. Therefore, the $l$ extent analysis values for each object can be obtained. These values are denoted as:

The two sides of the fuzzy number (left and right) are represented by $t\left(y\right) \mathrm{and} u\left(y\right)$. These numbers are denoted by the letters:

$$B_{k_i}^1,B_{k_i}^2,\dots ,B_{k_i}^m, i=1,2,\dots ,n$$

(3)

where $B_k^j\left(j=1,2,\dots ,m\right)$ are the TFNs.

The $i^{th}$ synthetic fuzzy extent value of the object is specified as:

$$Q_i={\displaystyle \sum }_{j=1}^m{B}_{k_i}^j\otimes {\left[{\displaystyle \sum }_{i=1}^n{\displaystyle \sum }_{j=1}^m{B}_{k_i}^j\right]}^{-1}$$

(4)

The possibility that $B_1\ge B_2$ is defined as:

$$V\left(B_1\ge B_2\right)=su{p}_{y\ge x}\left[min\left({\mu }_{B_1}\left(x\right),\right),\left({\mu }_{B_2}\left(y\right)\right)\right]$$

(5)

We have $V\left(B_1\ge B_2\right)=1$ when the pair $\left(x,y\right)$ occurs with $x\ge y$ and ${\mu }_{B_1}\left(x\right)={\mu }_{B_2}\left(y\right)$.

Because $B_1$ and $B_2$ are fuzzy convex numbers, we have:

$$\left(B_1\ge B_2\right)=1, if l_1\ge l_2$$

(6)

and

$$V\left(B_2\ge B_1\right)=hgt\left(B_1◠B_2\right)$$

(7)

where $d$ is the ordinate of the highest point of intersection D between ${\mu }_{B_1}$ and ${\mu }_{B_2}$.

The ordinate of point D is obtained by multiplying $B_1=(o_1,h_1$,$g_1)$ and $B_2=(o_2,h_2$,$g_2)$:

$$V\left(B_2\ge B_1\right)=hgt\left(B_1◠B_2\right)=\frac{l_1-g_2}{\left(h_2-g_2\right)-\left(h_1-o_1\right)}$$

(8)

We must calculate the values of $V\left(B_1\ge B_2\right)$ and $V\left(B_2\ge B_1\right)$ to compare $B_1$ and $B_2$.

The chance of a convex fuzzy number being bigger than $k$ convex fuzzy numbers $B_i\left(i=1,2,\dots k\right)$ is computed as:

$$\left(L\ge B_1,B_2,\dots ,B_k\right)=V\left[\left(B\ge B_1\right) and \left(B\ge B_2\right) \right]$$

(9)

and $(B\ge B_k$) = min V (B $\ge B_i), i=1,2,\dots , k$.

Taking the premise that:

$$d^{\prime }\left(J_i\right)=minV\left(Q_i\ge Q_k\right)$$

(10)

The weight vector for $k=1,2,\dots n \mathrm{and} k\#i$ is calculated as follows:

$$W^{\prime }={\left(d^{\prime }\left(J_1\right),d^{\prime }\left(J_2\right),\dots d^{\prime }\left(J_n\right)\right)}^T,$$

(11)

where $J_i$ are the $n$ elements.

The vectors of the normalized weights are displayed as:

$$W={\left(d\left(J_1\right),d\left(J_2\right),\dots .,d\left(J_n\right)\right)}^T$$

(12)

where $W$ is a nonfuzzy number.

3.2.2. Weighted Sum Method of Evaluation for Products

The weighted sum model (WSM) is one of the most popular and effective multicriteria decision-making techniques for evaluating multiple options in different criteria. There are options and $\mathit{c}$ criteria to consider first. The importance of the criteria ${\mathit{x}}_{\mathit{s}\mathit{c}}$ is then defined ${\mathit{w}}_{\mathit{c}}$, while the performance level for the option $\mathit{s}$ is examined and criterion $\mathit{c}$ is defined. Finally, the relative relevance of the alternative $\mathit{y}$ is defined by ${\mathit{L}}_{\mathit{y}}^{\left(\mathit{1}\right)}$ [22]:

$${\mathit{L}}_{\mathit{y}}^{\left(\mathit{1}\right)}={\displaystyle \sum }_{\mathit{c}=\mathit{1}}^{\mathit{n}}{\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}{\mathit{w}}_{\mathit{c}}$$

(13)

For each initial criteria value, the linear normalization is calculated as follows:

$${\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}=\frac{{\mathit{x}}_{\mathit{s}\mathit{c}}}{\mathit{m}\mathit{a}{\mathit{x}}_{\mathit{s}}{\mathit{x}}_{\mathit{s}\mathit{c}}}$$

(14)

when the $\mathit{m}\mathit{a}{\mathit{x}}_{\mathit{s}}{\mathit{x}}_{\mathit{s}\mathit{c}}$ cost is more important than the value or:

$${\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}=\frac{\mathit{m}\mathit{i}{\mathit{n}}_{\mathit{s}}{\mathit{x}}_{\mathit{s}\mathit{c}}}{{\mathit{x}}_{\mathit{s}\mathit{c}}}$$

(15)

when the $\mathit{m}\mathit{i}{\mathit{n}}_{\mathit{s}\mathit{c}}{\mathit{x}}_{\mathit{s}\mathit{c}}$ is more important than the value.

The weighted product model (WPM) is another strategy that is widely employed when evaluating numerous options $\mathit{y}$ based on their total relative value ${\mathit{L}}_{\mathit{y}}^{\left(2\right)}$:

$${\mathit{L}}_{\mathit{y}}^{\left(2\right)}={{\displaystyle \prod }}_{\mathit{c}=1}^{\mathit{n}}{\left({\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}\right)}^{{\mathit{w}}_{\mathit{c}}}$$

(16)

The weights of the total relative importance are then evenly divided between the WSM and WPM values for the total score in order to include both methodologies in the further analysis of the importance of the options:

$${\mathit{L}}_{\mathit{y}}=0.5{\mathit{L}}_{\mathit{y}}^{\left(1\right)}+0.5{\mathit{L}}_{\mathit{y}}^{\left(2\right)}$$

(17)

The coefficients that determine the WSM and WPM can then be further altered to adapt adequately to the problem, based on the study above and the further evaluation of the accuracy and effectiveness of the decision-making. The weighted aggregated sum product assessment method was used to rank the options in this study and it involved changing the coefficients:

$${\mathit{L}}_{\mathit{y}}=\mathit{\lambda }{\displaystyle \sum }_{\mathit{c}=1}^{\mathit{n}}{\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}{\mathit{w}}_{\mathit{c}}+\left(1-\mathit{\lambda }\right){{\displaystyle \prod }}_{\mathit{j}=1}^{\mathit{n}}{\left({\overline{\mathit{x}}}_{\mathit{s}\mathit{c}}\right)}^{{\mathit{w}}_{\mathit{c}}}$$

(18)

4. Case Study

The industrial revolution 4.0 is the core foundation for the future development of the logistics industry. It is not only involved in solving logistics problems for large companies and enterprises but also for start-ups that can apply and offer breakthrough solutions for each stage of supply, both in general and for logistics in particular. Thanks to the industrial revolution 4.0, companies and businesses can take advantage of opportunities to shorten order fulfillment times and satisfy customer demands. In the field of logistics, the industrial revolution 4.0 contributes to reducing delivery times, transportation costs and communication costs, thereby optimizing all business costs. At the same time, it helps to make the logistics systems and supply chains of companies and enterprises more transparent. However, as well as the positive aspects, Vietnam’s logistics industry also faces significant challenges in this transformation toward the 4.0 trend. Logistics has been operating with many shortcomings, such as high costs, difficult management and operations, difficult tracking of the status of goods, easy fraud and mainly manual operation. Therefore, it is very important to improve these difficulties within the logistics industry. One of the current priority options is the application of blockchain technology. Thanks to blockchain technology, the logistics industry could solve a series of its outstanding inadequacies. The workflow of blockchain technology in the logistics industry is shown in Figure 4.

With the scale of its dozens of subsidiaries and joint venture companies and especially with its significant contributions to the country’s economy, A Logistics JCS has been honored as being among the top of Vietnam’s leading logistics enterprises. Now firmly entering a new era of development, A Logistics JSC continues to consolidate its leading position and improve its core competencies: strengthening its team of solid experts, professional skills and enthusiasm; accelerating the expansion of the network, scale and scope of its service provision; strengthening the application of modern technology; and improving service quality to optimize business efficiency and provide outstanding value for its customers. Currently, the company is implementing the project of applying blockchain technology to the entire operation process. There are many blockchain development services providers and it is essential to choose the investment in this technology that suits the characteristics and purposes of the logistics organization in question. A Logistics JSC considered four blockchain logistics services providers (Block01, Block02, Block03 and Block04). The complexity of the criteria for choosing a suitable blockchain services provider turns the problem into a multicriteria decision-making problem. In this work, the author proposed an MCDM model for the assessment of blockchain development services providers for logistics organizations under uncertain environmental conditions. All criteria that affected the evaluation and selection process were chosen by industry experts and following a literature review. The list of criteria is shown in

Table 1. The evaluation criteria.

No	Criteria	Symbol
1	Costs of investment	BS01
2	Institution-based trust	BS02
3	Infrastructure facility	BS03
4	Compatibility	BS04
5	Availability of specific blockchain tools	BS05
6	Top management support	BS06
7	Capability of human resources	BS07
8	Government policy and support	BS08
9	Security and privacy	BS09
10	Ease of being tried and observed	BS10
11	Firm size	BS11

.

The weight of each criterion was calculated in the first stage of this study based on the analytic hierarchy process (AHP). The weights of the criteria are shown in

Table 2. The weights of all criteria.

Criteria	Fuzzy Sum of Each Row			Fuzzy Synthetic Extent			Degree of Possibility (Mi)	Normalization
BS01	9.9319	13.8996	18.5404	0.0544	0.1027	0.1879	0.8739	0.1097
BS02	9.2419	12.7572	17.1390	0.0507	0.0943	0.1737	0.6732	0.0845
BS03	11.1086	15.8349	21.5472	0.0609	0.1170	0.2183	0.9795	0.1230
BS04	9.0350	12.2780	16.5793	0.0495	0.0907	0.1680	0.6493	0.0815
BS05	12.1776	16.2651	21.2239	0.0668	0.1202	0.2151	1.0000	0.1255
BS06	6.3276	8.3802	11.5599	0.0347	0.0619	0.1171	0.4141	0.0520
BS07	6.7383	9.0654	12.3912	0.0369	0.0670	0.1256	0.4584	0.0576
BS08	9.6321	13.5526	18.2118	0.0528	0.1002	0.1845	0.8546	0.1073
BS09	9.6617	13.1833	17.4325	0.0530	0.0974	0.1766	0.8283	0.1040
BS10	7.5017	10.2407	14.1885	0.0411	0.0757	0.1438	0.6337	0.0796
BS11	7.3278	9.8572	13.6212	0.0402	0.0728	0.1380	0.6008	0.0754

.

Then, these weights were input into the WASPAS model to rank the various blockchain logistics services providers. The normalized matrix, the weighted normalized matrix and the exponentially weighted matrix are shown in

Table 3. The normalized matrix.

Criteria	Alternatives
	Block01	Block02	Block03	Block04
BS01	0.8750	1.0000	1.0000	1.0000
BS02	0.8750	0.7500	1.0000	0.6250
BS03	0.8750	0.7500	1.0000	0.6250
BS04	1.0000	0.8750	0.7500	1.0000
BS05	0.8750	1.0000	0.7500	1.0000
BS06	1.0000	1.0000	0.8750	1.0000
BS07	1.0000	0.8889	0.6667	0.5556
BS08	1.0000	0.8750	0.7500	0.5000
BS09	1.0000	0.7143	1.0000	0.5714
BS10	1.0000	0.8333	0.8333	1.0000
BS11	1.0000	0.6667	0.7778	0.6667

,

Table 4. The weighted normalized matrix.

Criteria	Alternatives
	Block01	Block02	Block03	Block04
BS01	0.0960	0.1097	0.1097	0.1097
BS02	0.0739	0.0634	0.0845	0.0528
BS03	0.1076	0.0922	0.1230	0.0768
BS04	0.0815	0.0713	0.0611	0.0815
BS05	0.1098	0.1255	0.0942	0.1255
BS06	0.0520	0.0520	0.0455	0.0520
BS07	0.0576	0.0512	0.0384	0.0320
BS08	0.1073	0.0939	0.0805	0.0536
BS09	0.1040	0.0743	0.1040	0.0594
BS10	0.0796	0.0663	0.0663	0.0796
BS11	0.0754	0.0503	0.0587	0.0503

and

Table 5. The exponentially weighted matrix.

Criteria	Alternatives
	Block01	Block02	Block03	Block04
BS01	0.9855	1.0000	1.0000	1.0000
BS02	0.9888	0.9760	1.0000	0.9611
BS03	0.9837	0.9652	1.0000	0.9438
BS04	1.0000	0.9892	0.9768	1.0000
BS05	0.9834	1.0000	0.9645	1.0000
BS06	1.0000	1.0000	0.9931	1.0000
BS07	1.0000	0.9932	0.9769	0.9667
BS08	1.0000	0.9858	0.9696	0.9283
BS09	1.0000	0.9656	1.0000	0.9435
BS10	1.0000	0.9856	0.9856	1.0000
BS11	1.0000	0.9699	0.9812	0.9699

.

In this study, the author proposed a two-stage fuzzy multicriteria decision-making model for the assessment of blockchain development services providers for logistics organizations. The weight of each criterion was calculated based on the analytic hierarchy process (AHP) in the first stage and then those weights were input into the WASPAS model to rank the four potential alternatives. As can be seen from the results in Figure 5 and

Table 6. The values of the coefficients.

Alternatives	Qi1	Qi2	Qi
Block01	0.9447	0.9855	0.9651
Block02	0.8500	1.0000	0.9250
Block03	0.8657	1.0000	0.9329
Block04	0.7733	1.0000	0.8866

, the ranking list was BLOCK01, BLOCK03, BLOCK02 and BLOCK04 with scores of 0.9651, 0.9329, 0.9250 and 0.8866, respectively. Thus, BLOCK01 was the optimal alternative.

5. Conclusions

Today’s logistics network is complex and inefficient due to a lack of visibility and outdated documentation systems. Blockchain technology combats these inefficiencies by allowing companies to track products in real time and store relevant data on tamper-proof ledgers. Blockchain technology stores information on a shared ledger that is viewable by all authorized parties, so companies can facilitate administrative processes as well. Choosing a blockchain logistics services provider is a very important task that can affect the performance of a logistics company. The decision-maker must consider many factors in the process of evaluating and selecting the optimal option for their company. In this work, the author proposed a fuzzy multicriteria decision-making model for the assessment of blockchain development services providers for logistics organizations. In the first stage, an FAHP model was used to calculate the weight of each criterion and then those weights were input into the WASPAS model to rank the various blockchain logistics services providers. The proposed model is the first selection model for blockchain logistics services provider in Vietnam that uses expert interviews and literature reviews. This is also the first work to utilize a combination of the FAHP and WASPAS models. The contribution of this work is the provision of useful guidelines for the evaluation and selection of blockchain logistics services providers for the logistics industry and other industries.

Although a consistency check of the fuzzy AHP model was performed in the present study, the inconsistency in the pairwise comparison matrix should not be neglected. This was a limitation of this study. This inconsistency could occur in other problems in practice. The weighted aggregated sum product assessment (WASPAS) can overcome this drawback as it reduces the burden on decision-makers by requiring fewer pairwise comparisons. Future research could extend the application of fuzzy numbers to the development of new MCDM models to solve decision-making problems in other fields and industries. Comparison studies could also be conducted to evaluate the performance of fuzzy MCDM models in comparison to other extensions of MCDM models.