Multi objective optimization model and algorithm of logistics service transaction matching on blockchain platform

Abstract

The logistics service transaction matching of the blockchain platform is an important part of the transaction. Firstly, the evaluation index system of the logistics service transaction satisfaction degree in the blockchain platform is constructed. According to the evaluation index of the logistics service users and providers, the actual situation of the logistics service transaction and the principle of maximizing the satisfaction degree of the logistics service users and providers, we set the expectation range of both sides and do data standardization, design the calculation method of satisfaction information amount for different expectation data types, build a multi-objective optimization model for logistics service transaction matching on blockchain platform, and seek the matching solution that makes all service users and providers the most satisfaction. Considering the large amount of data in blockchain, the concurrent design of NSGA-II algorithm based on Hadoop is carried out. Through the calculation of the example, in the matching of multi-objective optimization model of logistics service transaction matching on the blockchain platform, when the users and providers of logistics service are more satisfied about the indicators with each other, and then it gets more satisfactory matching results.

Keywords

Blockchain platform logistics service satisfaction transaction matching model NSGA-II algorithm

1. Introduction

In February 2018, the U.S. House of Representatives held many hearings about blockchain, elevating the “blockchain faith” to the US national strategy. On November 7, 2019, the British Parliament established a cross-party blockchain group global expert committee to jointly promote the development of blockchain with British MPs. On October 24, 2019, during the 18th collective study session of the Political Bureau of the CPC Central Committee, General Secretary Xi Jinping emphasized that “accelerating the innovation of blockchain technology and accelerating Industrial innovation.” This shows that China has raised the development of blockchain to the height of national strategy.

With the gradual establishment of logistics public information platforms, the network freight platforms, multimodal transport service platforms, and port public service platforms, the completed volume of the logistics service transactions will increase on these platforms. The logistics information platform is mainly to reduce customer risks, but it brings about many problems such as transaction lag, increased fees, and transaction information leakage. Because blockchain technology has the characteristics of decentralization and traceability, the integration of blockchain and logistics transactions can enable encrypted storage of data that cannot be tampered with in transactions. It is helpful to solve the problems of the lack of information and trust between both parties in logistics transactions to reduce corporate risks. It is a topic problem that needs to be broken about how to apply blockchain technology to logistics service transaction matching to ensure transaction security and convenience, while achieving intelligent matching optimization.

Blockchain technology is a distributed network data management technology. It uses cryptographic data and distributed consensus protocol to ensure the security of network transmission and access. It realizes multi-party data maintenance, cross verification, consistent throughout the network, and is not easy to tamper. As an important evolution of the new generation of information and communication technology, blockchain provides new ideas for the management and value release of data elements, and provides a new way for the establishment of a trusted cooperation network of cross industry entities. It is expected to play a more and more important role in the global economic recovery and the development of digital economy after the epidemic.

Blockchain can be divided into public blockchain and private blockchain. The validation process in blockchains can be done either by authorized users with permissioned access in private blockchains or implemented by unauthorized users with rewarding computer utilization in public blockchain technologies [1]. The public blockchains usually allows virtually anybody to freely interact during transactions, with or without prior knowledge of the identity of the interacting parties. On the other hand, there is sufficient prior knowledge of the identities of interacting parties in private blockchain systems during transactions. The public blockchains can also be differentiated from private blockchains in terms of selling dispositions; Public blockchains can assist firms to save cost and time while private blockchains can aid in the disintermediation of traditional intermediaries (e.g. banks) during business transactions. The freight logistics sector can utilize private blockchain systems for more transaction privacy critical for sensitive data while using the public ledger for data that require a high trust level and a substantial amount of computational power that is necessary to maintain distributed ledger on a larger scale [2].

In the research of logistics service transaction based on blockchain technology, many scholars put forward new ideas. Podgorelec et al. [1] considered a machine learning-based method, which introduces automated signing of blockchain transactions, while including also a personalized identification of anomalous transactions. Afterwards, Yang [5] proposed an associated ring signature method suitable for blockchain electronic currency transactions. They also find a block chain based electronic currency trading system security model. And the model not only ensures the anonymity of the system, but also has certain traceability. Choi and Luo [17] proposed how the blockchain technology can be applied to facilitate the implementation of mean-variance risk analysis for global supply chain operations. Meanwhile, Yang gave the maritime shipping blockchain-based digitalization pointed out the future improvement directions in the blockchain technology [5]. After, Yadav and Singh study the use of BC technology [16].

Blockchain technology will be widely used in various fields in the future, many experts put forward their own views Kshetri evaluated blockchain’s roles in strengthening cybersecurity and protecting privacy [8]. He provided a detailed analysis and description of blockchain’s roles in tracking the sources of insecurity in supply chains related to IoT devices. After that, Tang and Veelenturf discussed the potential use of emerging blockchain [4]. By using new technologies (additive manufacturing, advanced robotics, artificial intelligence, driverless, blockchain, UAV, Internet of things), many companies are developing information physical systems that can change the competitive landscape. They found that logistics and transportation services play a strategic role in developing economic, environmental and social values. In the same year, Issaoui et al. defined the various applications of Blockchain in Smart Logistics [20]. López and Farooq demonstrated the performance of BSMD on a 370 nodes blockchain running on heterogeneous [6]. Then, Shen et al. examined the value of blockchain for disclosing secondhand product quality in a supply chain [2]. In addition. Christidis and Devetsikioti proposed that blockchain-IoT combination is powerful and can cause significant transformations across several industries [9]. One year later, Wang et al. put forward the block chain technology to modify the framework scheme and preliminary design of the architecture [8]. Then, Li et al. proposed a blockchain-enabled workflow operating system (Bc-WfOS) to centrally share heterogeneous logistics resources with different customers [13]. As well as Choi and Luo considered the implementation of blockchain to help and identify the situation in which blockchain helps enhance social welfare but brings harm to supply chain profitability [17]. Recently, Orji et al. examined a technology- organization- environment (TOE) theoretical framework of critical factors that influence the successful adoption of blockchain technologies in the freight logistics industry and prioritize them using the analytic network process (ANP) [7]. Lately, Cheung et al. thought that blockchain technologies are still in their infancy in the transport and logistics secto [10]. And Niu et al. found that block-chain is a double edged sword for the MNF [3]. It increases the MNF’s wholesaling profit, but reduces the MNF’s retailing profit and tax planning benefit.

In the transaction process of blockchain-based logistics service, the first step is the transaction matching link. Both of the users and provides can enter the smart contract process after reaching a satisfactory match under certain rules. There are studies on the single-objective model method that uses the weight of the price utility function of both parties in the logistics service transaction matching link under the blockchain [20]. The matching attribute is a single price factor, which cannot fully reflect the demand and supply of logistics service users and providers. This paper studies the establishment of a satisfaction evaluation index system for both parties in logistics service transactions, using information axiom theory, setting the expected value range of both parties and standardizing data, establishing a logistics service transaction matching optimization model, designing a Hadoop-based NSGA-II parallel multi-objective optimization algorithm, and seeking a matching plan that maximizes the satisfaction of all service users and providers.

2. Establishment of an evaluating satisfaction degree index system of logistics service transactions on blockchain platform

In the logistics service transaction based on block chain, there is an intelligent matching of single or multiple objectives between logistics service users and logistics service providers. The relevant description information of the two parties is respectively stored in the block chain block. The block chain platform provides the real information of the two parties according to the objective historical database, and carries out the transaction matching according to the expected information of the two parties, so as to achieve the expected result of the two parties.

As the logistics service users and providers have different priorities, preferences and perspectives for the same transaction, they each have various attribute demands, and both of them have a great impact on the completion of the logistics transaction. Therefore, it is necessary to establish an index system for the satisfaction evaluation of logistics service transactions on the block chain platform. Real number, interval number and triangular fuzzy number were used to measure quantitative index, while intuitionistic fuzzy number was used to measure qualitative index.

2.1 The satisfaction evaluation index system of logistics service users to the providers

We research some literatures and combine with the actual situation, then we select corresponding evaluation index system, these are 9 indicators from three dimensions of operational level, enterprise cooperation trend and service ability, and the evaluation index system is shown in Table 1.

Table 1
The evaluation index system of logistics service users to providers

Dimension	Evaluation index	Representation	Indicator type
Operational level	market share	Verbal/intuitive fuzzy number	Benefit type
	Value-added service capabilities	Verbal/intuitive fuzzy number	Benefit type
	Technical soft power	Verbal/intuitive fuzzy number	Benefit type
Enterprise cooperation trend	Corporate reputation	Verbal/intuitive fuzzy number	Benefit type
	Collaborative spirit	Verbal/intuitive fuzzy number	Benefit type
Service abilities	price	Triangular fuzzy number	Cost type
	Outlet layout coverage rate	interval number	Benefit type
	Cargo time efficiency	Real number	Benefit type
	Cargo damage rate	interval number	Cost type

2.2 The satisfaction evaluation index system of logistics service providers to users

Six indicators are selected from the two dimensions of current value and potential value, and the evaluation index system is shown in Table 2.

Table 2
Evaluation index system of logistics service providers to users

Dimension	Evaluation index	Representation	Indicator
Current value	purchase amount	Triangular fuzzy number	Benefit type
	Payment cycle	Number of intervals	Cost type
	Business scale	Verbal/intuitive fuzzy number	Benefit type
	Bank credit rating	Real number	Benefit type
Potential value	Business growth rate	Verbal/intuitive fuzzy number	Benefit type
	Possibility of cross-selling	Verbal/intuitive fuzzy number	Benefit type

3. Construction of the logistics service transaction on platform to match the multi-objective optimization model

3.1 Problem description

Assuming that multiple logistics services users and providers seek transactions on the block chain platform at the same time, and we put forward their respective expectations to the platform. The platform objectively provides the actual information on both parties to match based on the database and historical information to achieve mutual satisfaction. According to the expected value range, historical actual data and the evaluation index, when we calculate the satisfaction degree, the information amount of the opposite side is calculated, and then the satisfaction information value is summed to obtain the total satisfaction information value.

Set all logistics service users to be $U=\left\{{U_{1},U_{2},U_{3},\ldots,U_{m}}\right\}$ , where $U_{i}$ represents the $i$ -th logistics service user; Set all logistics service providers be $P=\left\{{P_{1},P_{2},P_{3},\ldots,P_{n}}\right\}$ , where $P_{j}$ represents the $j$ -th logistics service provider. Set the evaluation index of logistics service users to providers is $E=\left\{{e_{1},e_{2},e_{3},\ldots,e_{q}}\right\}$ , where $e_{h}$ represents the $h$ -th evaluation index; Set the evaluation index of logistics service providers to users is $R=\left\{{r_{1},r_{2},r_{3},\ldots,r_{s}}\right\}$ , where $r_{g}$ is the $g$ -th evaluation index.

Let the evaluation index of logistics service users and providers for each other to be $E=E_{1}\cup E_{2}\cup E_{3}\cup E_{4}$ , $R=R_{1}\cup R_{2}\cup R_{3}\cup R_{4}$ , where $E_{i}\cap E_{j}=\emptyset$ , $R_{i}\cap R_{j}=\emptyset\left({i,j=1,2,3,4;i\neq j}\right)$ , $E_{i},R_{i}\left({i=1,2,3,4}\right)$ indicate that the evaluation information is an index set of real numbers, interval numbers, triangular fuzzy numbers and intuitive fuzzy numbers. The benefit type of $E_{i}$ is represented by $E_{i}^{a}$ , and the cost type is represented by $E_{i}^{b}$ . At the same time, it needs to satisfy $E_{i}^{a}\cap E_{i}^{b}=\emptyset$ , $E_{i}^{a}\cup E_{i}^{b}=E_{i}$ ; Similarly, the benefit type of $R_{i}$ is represented by $R_{i}^{a}$ , the cost type is expressed by $R_{i}^{b}$ , and it needs to satisfy $R_{i}^{a}\cap R_{i}^{b}=\emptyset$ , $R_{i}^{a}\cup R_{i}^{b}=R_{i}$ . Hesitation degree and non-membership degree are added to intuitive fuzzy numbers to ensure that the evaluation information is not lost and the decision-makers’ thoughts are correctly expressed.

Suppose the logistics services user $U_{i}$ has an expected value range of $e_{h}$ as:

$\displaystyle T_{ih}=\left\{{{\begin{array}[]{ll}d_{ih}&\textit{if}\ e_{h}\in E% _{1}\\ \left[{t^{\prime}_{ih},t^{\prime\prime}_{ih}}\right]&\textit{if}\ e_{h}\in E_{% 2}\\ \left({a_{ih},b_{ih},c_{ih}}\right)&\textit{if}\ e_{h}\in E_{3}\\ \lambda_{ih},\partial_{ih}&\textit{if}\ e_{h}\in E_{4}\\ \end{array}}}\right.$ (1)

Through the objective evaluation of various historical data onto $P_{j}$ in the blockchain platform, the range of $P_{j}$ in the indicator $e_{h}$ is obtained:

$\displaystyle H_{jh}=\left\{{{\begin{array}[]{ll}g_{jh}&\textit{if}\ e_{h}\in E% _{1}\\ \left[{h^{\prime}_{jh},h^{\prime\prime}_{jh}}\right]&\textit{if}\ e_{h}\in E_{% 2}\\ \left({o_{jh},q_{jh},l_{jh}}\right)&\textit{if}\ e_{h}\in E_{3}\\ \mu_{jh},\omega_{jh}&\textit{if}\ e_{h}\in E_{4}\\ \end{array}}}\right.$ (2)

Set the logistics service provider $P_{j}$ have an expected value range of the index $r_{g}$ as:

$\displaystyle\bar{T}_{jg}=\left\{{{\begin{array}[]{ll}\bar{d}_{jg}&\textit{if}% \ r_{g}\in R_{1}\\ \left[{\bar{t}^{\prime}_{jg},\bar{t}^{\prime\prime}_{jg}}\right]&\textit{if}\ % r_{g}\in R_{2}\\ \left({\bar{a}_{jg},\bar{b}_{jg},\bar{c}_{jg}}\right)&\textit{if}\ r_{g}\in R_% {3}\\ \bar{\lambda}_{jg},\bar{\partial}_{jg}&\textit{if}\ r_{g}\in R_{4}\\ \end{array}}}\right.$ (3)

Similarly, through the objective evaluation of various historical data of $U_{i}$ in the blockchain platform, the range of $U_{i}$ in the indicator $r_{g}$ is:

$\displaystyle\bar{H}_{ig}=\left\{{{\begin{array}[]{ll}\bar{g}_{ig}&\textit{if}% \ r_{g}\in R_{1}\\ \left[{\bar{h}^{\prime}_{ig},\bar{h}^{\prime\prime}_{ig}}\right]&\textit{if}\ % r_{g}\in R_{2}\\ \left({\bar{o}_{ig},\bar{q}_{ig},\bar{l}_{ig}}\right)&\textit{if}\ r_{g}\in R_% {3}\\ \bar{\mu},\bar{\omega}_{ig}&\textit{if}\ r_{g}\in R_{4}\\ \end{array}}}\right.$ (4)

It is assumed that each $U_{i}$ can trade with $\varphi_{i}$ logistics service providers at most, and each $P_{j}$ can trade with $\theta_{j}$ logistics service users at most. When $\varphi_{i}=1$ and $\theta_{j}=1$ , It is an one-to-one transaction; When any of the $\varphi_{i}$ and $\theta_{j}$ equals 1, and the other is greater than 1, it is an one-to-many transaction; When they are both greater than 1, it is a many-to-many transaction.

3.2 Data standardization

According the reference documents of data standardization, the expected range of logistics service users to providers is represented by $T_{ih}$ .

If $T_{ih}=d_{ih}\left({e_{h}\in E_{1}}\right)$ , and the indicator is a real number indicator:

$\displaystyle\delta_{ik}=\left\{{{\begin{array}[]{ll}\frac{d_{ih}}{d_{\textit{% MAXh}}},&\textit{if}\ e_{h}\in E_{1}^{b}\\ 1-\frac{d_{ih}}{d_{\textit{MAXh}}},&\textit{if}\ e_{h}\in E_{1}^{c}\\ \end{array}}}\right.$ (5)

Among them, $d_{\textit{MAXh}}=\max\left\{{d_{ih}\left|{i=1,2,\ldots,m}\right.}\right\}$ .

When $T_{ih}=\left[t^{\prime}_{ih},t^{\prime\prime}_{ih}\right]\left({e_{h}\in E_{2}% }\right)$ , and the indicator is an interval number indicator:

$\displaystyle\delta_{ik}=\left\{\begin{array}[]{ll}\left[\frac{t^{\prime}_{ih}% }{t^{\prime\prime}_{\textit{MAXh}}},\frac{t^{\prime\prime}_{ih}}{t^{\prime% \prime}_{\textit{MAXh}}}\right],&\textit{if}\ e_{h}\in E_{2}^{b}\\ \left[1-\frac{t^{\prime\prime}_{ih}}{t^{\prime\prime}_{\textit{MAXh}}},1-\frac% {t^{\prime}_{ih}}{t^{\prime\prime}_{\textit{MAXh}}}\right],&\textit{if}\ e_{h}% \in E_{2}^{c}\\ \end{array}\right.$ (6)

Among them, $t^{\prime\prime}_{\textit{MAXh}}=\max\left\{t^{\prime\prime}_{ih}|i=1,2,\ldots% ,m\right\}$ .

When $T_{ih}=\left({a_{ih},b_{ih},c_{ih}}\right)\left({e_{h}\in E_{3}}\right)$ , the indicator is a triangular fuzzy number indicator:

$\displaystyle\delta_{ik}=\left\{{{\begin{array}[]{ll}\left[{\frac{a_{ih}}{c_{% \textit{MAXh}}},\frac{b_{ih}}{c_{\textit{MAXh}}},\frac{c_{ih}}{c_{\textit{MAXh% }}}}\right],&\textit{if}\ e_{h}\in E_{2}^{b}\\ \left[{1-\frac{c_{ih}}{c_{\textit{MAXh}}},1-\frac{b_{ih}}{c_{\textit{MAXh}}},1% -\frac{a_{ih}}{c_{\textit{MAXh}}}}\right],&\textit{if}\ e_{h}\in E_{2}^{c}\\ \end{array}}}\right.$ (7)

Among them, $c_{\textit{MAXh}}=\max\left\{c_{ih}|{i=1,2,\ldots,m}\right\}$ .

When $T_{ih}=\lambda_{ih},\partial_{ih}\left({e_{h}\in E_{4}}\right)$ , the indicator is an intuitive fuzzy number indicator. The number between 0 and 1 shouldn’t be standardized, but can be used directly.

After data standardization is completed, it represents that the cost-based indicator is converted to the profit-based indicator.

3.3 Satisfaction information calculation

3.3.1 Calculation of real number satisfaction information volume

Logistics service transaction matching parties use precise values to express their expected and actual value metric types. The logistics services user $U_{i}$ expects the index $e_{h}\in E_{1}$ to be $d_{ih}$ , and $P_{j}$ is $g_{jh}$ at the actual level of the index $e_{h}\in E_{1}$ . When the indicator is set as the profit indicator $e_{h}\in E_{1}^{b}$ . For users, the larger the actual value of the target the better, if it can exceed its expected value, it is even better. When the actual value exceeds the expected value, it means that the maximum satisfaction is reached. And the maximum satisfaction is expressed as information equals zero. When there is a certain distance between the actual value of the logistics service user and the expected value, it means that it did not meet the requirements. The farther the distance, the more dissatisfaction. The calculation formula for the amount of user satisfaction information is shown as follows:

$\displaystyle I\left({T_{ih},H_{jh}}\right)=\left\{{{\begin{array}[]{ll}{\rm log% }_{2}e^{d_{ih}-g_{jh}}&\textit{if}\ d_{ih}>g_{jh}\\ 0&\textit{if}\ d_{ih}<g_{jh}\\ \end{array}}}\right.$ (8)

Similarly, the amount of information $P_{j}$ ’s satisfaction with $U_{i}$ when the indicator is $r_{g}\in R_{1}$ equals $I(\bar{T}_{jg},\bar{H}_{ig})$ :

$\displaystyle I(\bar{T}_{jg},\bar{H}_{ig})=\left\{\begin{array}[]{ll}\log_{2}e% ^{\bar{d}_{jg}-\bar{g}_{ig}}&\textit{if}\ \bar{d}_{jg}>\bar{g}_{ig}\\ 0&\textit{if}\ \bar{d}_{jg}<\bar{g}_{ig}\end{array}\right.$ (9)

3.3.2 Calculation of interval index satisfaction information volume

The attribute value is an interval number. If the actual range and the expected overlap, it means that the requirement is satisfied, and the information value reaches its minimum at this time; If it exceeds the expected value range. It means that the requirement not meets. At this time, the information value is the maximum, and it is impossible to measure who is better. Take the income indicator as an example, the expected range given by $U_{i}$ for the indicator $e_{h}$ is $\left[t^{\prime}_{ih},t^{\prime\prime}_{ih}\right]$ , and the actual range given by $P_{j}$ is $\left[h^{\prime}_{jh},h^{\prime\prime}_{jh}\right]$ , the higher the attribute value of the provider the better, and the closer to $t^{\prime}_{ih}$ is better which will make two parties better match. $U_{i}$ ’s calculation formula for the information amount of satisfaction with $P_{j}$ interval numbers indicator when $e_{h}\in E_{2}^{b}$ :

$\displaystyle I\left({T_{ih},H_{jh}}\right)=\left\{{{\begin{array}[]{ll}\log_{% 2}e^{\left[\frac{t^{\prime\prime}_{ih}-h_{jh}}{t^{\prime\prime}_{ih}-t^{\prime% }_{ih}}\right]}&\textit{if}\ t^{\prime\prime}_{ih}>h_{jh}\\ 0&\textit{if}\ t^{\prime\prime}_{ih}<h_{jh}\\ \end{array}}}\right.$ (10)

Among them, $h_{jh}$ is $\frac{h^{\prime}_{jh}+h^{\prime\prime}_{jh}}{2}$ .

Similarly, the information value $I\left(\bar{T}_{jg},\bar{H}_{ig}\right)$ for $U_{i}$ in the indicator $r_{g}\in R_{2}$ of $P_{j}$ is obtained.

3.3.3 Calculation of satisfaction information amount of triangular fuzzy number indicator volume

Assume that the expected range of the $U_{i}$ side of the logistics service user whose indicator $e_{h}$ is $\left({a_{ih},b_{ih},c_{ih}}\right)$ , and the $e_{h}$ at the actual level index of the logistics service demand sides $P_{j}$ is $\left({o_{jh},q_{jh},l_{jh}}\right)$ . The formula for the amount of information about the triangular fuzzy number index satisfaction is as follows:

$\displaystyle I\left({T_{ih},H_{jh}}\right)=\left\{{{\begin{array}[]{ll}\log_{% 2}e^{\frac{D_{c}}{S_{C}}}&\textit{if}\ a_{ih}>o_{jh}\ \textit{and}\ b_{ih}>q_{% jh}\\ 0&\textit{if}\ a_{ih}\leqslant o_{jh}\ \textit{and}\ b_{ih}\leqslant q_{jh}\\ \end{array}}}\right.$ (11)

Among them, $S_{c}$ is the actual scope of the provider of the logistics service; $D_{c}$ is the difference between the expected scope and the actual scope.

Similarly, $P_{j}$ is satisfied with the information value $I(\bar{T}_{jg},\bar{H}_{ig})$ for $U_{i}$ at the index $r_{g}\in R_{3}$ .

3.3.4 Calculation of intuitive fuzzy number index satisfaction information volume

The calculation formula for intuitive fuzzy cross-entropy includes non-membership and hesitation, which are more reasonable and scientific, and can calculate the amount of information about intuitionistic fuzzy number indicator satisfaction. $U_{i}$ ’s satisfaction with $P_{j}$ at $e_{h}\in E_{5}$ is $I\left({T_{ih},H_{jh}}\right)$ , the smaller the value, the smaller the difference between the logistics service transaction provider and the user in the indicator $U_{h}$ , the higher the matching degree, and the higher the amount of satisfaction information. Similarly, the satisfaction information amount $I(\bar{T}_{jg},\bar{H}_{ig})$ of $P_{j}$ to $U_{i}$ in the indicator $r_{g}\in R_{5}$ is obtained.

$\displaystyle D\left({T_{ih},H_{jh}}\right)=\lambda_{ih}{\rm log}_{2}\frac{% \lambda_{ih}}{\frac{1}{2}\left({\lambda_{ih}+\mu_{jh}}\right)}+\partial_{ih}{% \rm log}_{2}\frac{\partial_{ih}}{\frac{1}{2}\left({\partial_{ih}+\omega_{jh}}% \right)}+\mu_{jh}{\rm log}_{2}\frac{\mu_{jh}}{\frac{1}{2}\left({\lambda_{ih}+% \mu_{jh}}\right)}+\omega_{jh}{\rm log}_{2}\frac{\omega_{jh}}{\frac{1}{2}\left(% {\partial_{ih}+\omega_{jh}}\right)}$ (12)

$\displaystyle I\left({T_{ih},H_{jh}}\right)=\log_{2}e^{D^{\left({T_{ih},H_{jh}% }\right)}}$ (13)

For efficiency indicators, if $\lambda_{ih}<\mu_{jh}$ or $\lambda_{ih}=\mu_{jh}$ and $\partial_{ih}>\omega_{jh}$ , then $I\left({T_{ih},H_{jh}}\right)=0$ .

For language variable indicators, it is usually necessary to convert language information on fuzzy number information on calculation, that is, to convert language variable indicators into intuitionistic fuzzy numbers, which can be more fully to reflect the true intentions. This paper uses five-grained language phrases and intuitive fuzzy number conversion.

3.4 Model building

According to the satisfaction values $I\left({T_{ih},H_{jh}}\right)$ and $I\left({\bar{T}_{jg},\bar{H}_{ig}}\right)$ of both the logistics service user and the provider on each index, the overall satisfaction information value $I\left({U_{i},P_{j}}\right)$ and $I\left({U_{j},P_{i}}\right)$ are calculated.

The formula for calculating the overall satisfaction information volume of logistics services user $U_{i}$ and provider $P_{j}$ :

$\displaystyle I\left({U_{i},P_{j}}\right)=\sum^{q}_{h=1}I\left({T_{ih},H_{jh}}\right)$ (14)

The overall satisfaction degree of $U_{i}$ to $P_{j}$ is the degree of satisfaction from the overall expectation level of $U_{i}$ by $P_{j}$ under each index. The smaller the value of $I\left({U_{i},P_{j}}\right)$ , the smaller the total amount of information. It also shows that the closer the expected level of $U_{i}$ and the actual level of $P_{j}$ is, and the more satisfied the users of logistics services are, which means that $U_{i}$ is more satisfied with $P_{j}$ .

The formula for calculating the overall satisfaction information about logistics service provider $P_{j}$ and user $U_{i}$ :

$\displaystyle I\left({P_{j},U_{i}}\right)=\sum^{s}_{g=1}I\left({\bar{T}_{jg},% \bar{H}_{ig}}\right)$ (15)

A multi-objective optimization models for logistics service transaction matching is constructed as follows:

$\displaystyle\text{max}Z_{1}=\sum_{i=1}^{m}\sum^{n}_{j=1}\left({M_{1}-I\left({% U_{i},P_{j}}\right)}\right)x_{ij}$ (16) $\displaystyle\text{max}Z_{2}=\sum_{j=1}^{n}\sum^{m}_{i=1}\left({M_{2}-I\left({% P_{j},U_{i}}\right)}\right)x_{ij}$ (17)

$\displaystyle s.t.\sum^{m}_{i=1}x_{ij}\leqslant\theta_{j},j=1,2,\ldots,n$ (18) $\displaystyle\sum^{n}_{j=1}x_{ij}\leqslant\varphi_{i},i=1,2,\ldots,m$ (19) $\displaystyle x_{ij}=\left\{{0,1}\right\},i=1,2,\ldots,m,j=1,2,\ldots,n$ (20)

Among them:

$M_{1}=\max\left\{{I\left({U_{i},P_{j}}\right)}\right\},M_{1}$ is the maximum amount of information about the evaluation of logistics service users to providers. $M_{2}=\max\left\{{I\left({P_{j},U_{i}}\right)}\right\},M_{2}$ is the maximum amount of information about the evaluation of logistics service providers to users.

Instructions:

(10) means that the total amount of information about matching solutions to all logistics service users and providers is the smallest, and the satisfaction of all logistics service users is the maximum. (11) Means to maximize the satisfaction of all logistics service providers. (12)–(14) are constraints, where (12) means that the logistics service provider $P_{j}$ matches $\theta_{j}$ users at most; (13) means that the logistics service user $U_{i}$ matches $\phi_{i}$ providers at most; (14) is a decision variable $x_{ij}$ , which is a 0-1 variable, $x_{ij}=1$ means that the logistics service user $U_{i}$ matches the provider $P_{j}$ ; $x_{ij}=0$ means that the logistics service user $U_{i}$ does not match the provider $P_{j}$ .

4. Model algorithm design

4.1 Design of NSGA-II parallel multi-objective optimization algorithm based on Hadoop

In order to improve the computing efficiency of multi-objective optimization model, NSGA-II parallel multi-objective optimization algorithm is adopted in Hadoop framework. First, Hadoop cluster is built and Java running environment is configured; After successful debugging, NSGA-II parallelization is run on MapReduce computing engine; Data is interacted between Hadoop and block chain platform by reading and writing files; MapReduce separates separate data block files for parallel processing and computation through the Map process. Running the program to verify the effectiveness of the Hadoop based parallelization NSGA-II algorithm.

4.2 The example analysis

Combined with the actual situation to set the initial data at random in a reasonable range. The number of logistics service providers are 10, represented by $\left({P_{1},P_{2},P_{3}\ldots P_{10}}\right)$ , and the number of logistics service to consumers is 10, represented by $\left({U_{1},U_{2},U_{3}\ldots U_{10}}\right)$ .

4.2.1 Calculation of the satisfaction information value

The excepted information about consumer to provider on logistics service is shown in Table 3, $\left\{{e_{1},e_{2},e_{3},e_{4},e_{5},e_{6},e_{7},e_{8},e_{9}}\right\}$ are users’ expected evaluation indexes, respectively, the price (yuan/kg), outlet layout coverage, cargo time efficiency (per month), cargo damage rate (per month), market share, value-added service capabilities, technical of soft power, the enterprise reputation, collaborative spirit. The expected information on logistics service providers to consumers is shown in Table 4. $\left\{{e_{1},e_{2},e_{3},e_{4},e_{5},e_{6},e_{7},e_{8},e_{9}}\right\}$ are the evaluation indicators of providers’ expectations, which are purchase amount (ten thousand yuan), payment cycle (month), business scale, bank credit rating, business growth rate and possibility of cross-selling, respectively.

Table 3
Expected information about users to providers on logistics service

	$e_{1}$	$e_{2}$	$e_{3}$	$e_{4}$	$e_{5}$	$e_{6}$	$e_{7}$	$e_{8}$	$e_{9}$
$U_{1}$	(8,10,12)	(2,3)	0.94	(0.005,0.008)	(0.06,0.12)	(“M”,0.3)	(“M”,0.3)	(“G;”,0.3)	(“G”,0.3)
$U_{2}$	(7,9,11)	(3,4)	0.92	(0.006,0.008)	(0.05,0.1)	(“M”,0.1)	(“M”,0.1)	(“G”,0.1)	(“G”,0.1)
$U_{3}$	(14,16,18)	(1,2)	0.99	(0.002,0.004)	(0.15,0.2)	(“VG”,0.1)	(“VG”,0.1)	(“VG”,0.1)	(“VP”,0.1)
$U_{4}$	(10,12,14)	(1.5,3)	0.96	(0.004,0.006)	(0.1,0.15)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)
$U_{5}$	(8,10,12)	(2,3)	0.95	(0.005,0.007)	(0.08,0.12)	(“G”,0.1)	(“VG”,0.1)	(“G”,0.1)	(“G”,0.1)
$U_{6}$	(8,10,12)	(1,2)	0.96	(0.003,0.005)	(0.1,0.15)	(“VP”,0.1)	(“VP”,0.1)	(“VP”,0.1)	(“VG”,0.1)
$U_{7}$	(9,11,13)	(2,3)	0.92	(0.006,0.008)	(0.05,0.1)	(“VG”,0.1)	(“G”,0.1)	(“VG”,0.1)	(“VG”,0.1)
$U_{8}$	(10,12,14)	(1.5,3)	0.99	(0.002,0.004)	(0.15,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)
$U_{9}$	(7,9,11)	(2,3)	0.96	(0.004,0.006)	(0.1,0.15)	(“G”,0.1)	(“G”,0.1)	(“G”,0.1)	(“VG”,0.1)
$U_{10}$	(12,14,16)	(1,2)	0.95	(0.005,0.007)	(0.08,0.12)	(“VP”,0.1)	(“VP”,0.1)	(“VP”,0.1)	(“G”,0.2)

Table 4

Expected information about providers to users on logistics service

	$r_{1}$	$r_{2}$	$r_{3}$	$r_{4}$	$r_{5}$	$r_{6}$
$P_{1}$	(2.0,3.0)	(8.0,10.0,12.0)	(“M”,0.3)	4	(“M”,0.3)	(“M”,0.3)
$P_{2}$	(3.0,4.0)	(7.0,9.0,11.0)	(“M”,0.1)	5	(“M”,0.1)	(“M”,0.1)
$P_{3}$	(1.0,2.0)	(14.0,16.0,18.0)	(“VG”,0.1)	4	(“VG”,0.1)	(“VG”,0.1)
$P_{4}$	(1.5,3.0)	(10.0,12.0,14.0)	(“G”,0.2)	3	(“G”,0.2)	(“G”,0.2)
$P_{5}$	(2.0,3.0)	(8.0,10.0,12.0)	(“G”,0.1)	4	(“G”,0.1)	(“G”,0.1)
$P_{6}$	(1.0,2.0)	(8.0,10.0,12.0)	(“VG”,0.08)	5	(“VG”,0.1)	(“VP”,0.08)
$P_{7}$	(2.0,3.0)	(8.0,10.0,12.0)	(“G”,0.1)	4	(“G”,0.1)	(“G”,0.1)
$P_{8}$	(1.0,2.0)	(8.0,10.0,12.0)	(“VG”,0.08)	5	(“VP”,0.1)	(“VG”,0.18)
$P_{9}$	(1.0,2.0)	(14.0,16.0,18.0)	(“M”,0.1)	4	(“M”,0.1)	(“M”,0.1)
$P_{10}$	(1.5,3.0)	(10.0,12.0,14.0)	(“VG”,0.08)	3	(“VG”,0.1)	(“VG”,0.1)

The actual level of logistics transaction service users is shown in Table 5, and the actual level of logistics transaction service providers is shown in Table 6.

Table 5

The actual level of logistics service users

	$r_{1}$	$r_{2}$	$r_{3}$	$r_{4}$	$r_{5}$	$r_{6}$
$U_{1}$	(2.0,3.0)	(8.0,10.0,12.0)	(“M”,0.3)	4	(“M”,0.3)	(“M”,0.3)
$U_{2}$	(3.0,4.0)	(7.0,9.0,11.0)	(“M”,0.1)	5	(“M”,0.1)	(“M”,0.1)
$U_{3}$	(1.0,2.0)	(14.0,16.0,18.0)	(“VG”,0.1)	4	(“VG”,0.1)	(“VG”,0.1)
$U_{4}$	(1.5,3.0)	(10.0,12.0,14.0)	(“G”,0.2)	3	(“G”,0.2)	(“G”,0.2)
$U_{5}$	(2.0,3.0)	(8.0,10.0,12.0)	(“G”,0.1)	4	(“G”,0.1)	(“G”,0.1)
$U_{6}$	(1.0,2.0)	(8.0,10.0,12.0)	(“VP”,0.08)	5	(“VP”,0.08)	(“VP”,1.18)
$U_{7}$	(2.0,3.0)	(7.0,9.0,11.0)	(“VG”,0.1)	5	(“M”,0.1)	(“VG”,0.1)
$U_{8}$	(3.0,4.0)	(14.0,16.0,18.0)	(“G”,0.2)	4	(“VG”,0.1)	(“G”,0.2)
$U_{9}$	(1.0,2.0)	(10.0,12.0,14.0)	(“G”,0.1)	3	(“G”,0.2)	(“G”,0.1)
$U_{10}$	(1.5,3.0)	(8.0,10.0,12.0)	(“VP”,0.08)	4	(“G”,0.1)	(“VP”,1.18)

Table 6

The actual level of logistics service providers

	$e_{1}$	$e_{2}$	$e_{3}$	$e_{4}$	$e_{5}$	$e_{6}$	$e_{7}$	$e_{8}$	$e_{9}$
$P_{1}$	(8.0,10.0,12.0)	(2.0,3.0)	(0.94)	(0.005,0.008)	(0.06,0.12)	(“M”,0.3)	(“M”,0.3)	(“G”,0.3)	(“M”,0.3)
$P_{2}$	(7.0,9.0,11.0)	(3.0,4.0)	(0.92)	(0.006,0.008)	(0.05,0.1)	(“M”,0.1)	(“M”,0.1)	(“G”,0.1)	(“M”,0.1)
$P_{3}$	(14.0,16.0,18.0)	(1.0,2.0)	(0.99)	(0.002,0.004)	(0.15,0.2)	(“VG”,0.1)	(“VG”,0.1)	(“VG”,0.1)	(“VG”,0.1)
$P_{4}$	(10.0,12.0,14.0)	(1.5,3.0)	(0.96)	(0.004,0.006)	(0.15,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)
$P_{5}$	(8.0,10.0,12.0)	(2.0,3.0)	(0.95)	(0.005,0.007)	(0.08,0.12)	(“G”,0.1)	(“G”,0.1)	(“G”,0.1)	(“G”,0.1)
$P_{6}$	(8.0,10.0,12.0)	(1.0,2.0)	(0.98)	(0.003,0.005)	(0.1,0.15)	(“VP”,0.1)	(“VP”,0.1)	(“VP”,0.1)	(“VG”,0.08)
$P_{7}$	(8.0,10.0,12.0)	(2.0,3.0)	(0.95)	(0.005,0.007)	(0.08,0.12)	(“G”,0.1)	(“G”,0.1)	(“G”,0.1)	(“G”,0.1)
$P_{8}$	(8.0,10.0,12.0)	(1.0,2.0)	(0.98)	(0.003,0.007)	(0.1,0.15)	(“VP”,0.1)	(“VP”,0.1)	(“VG”,0.1)	(“VP”,0.18)
$P_{9}$	(10.0,12.0,14.0)	(2.0,3.0)	(0.99)	(0.004,0.006)	(0.05,0.1)	(“VG”,0.1)	(“VG”,0.1)	(“VG”,0.1)	(“VG”,0.1)
$P_{10}$	(8.0,10.0,12.0)	(1.0,2.0)	(0.96)	(0.005,0.007)	(0.15,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)	(“G”,0.2)

Calculation steps of satisfactory information value:

(1)

Normalize the data in Tables 3–6.

(2)

Calculate the satisfaction information value, and obtain the satisfaction information amount $I\left({U_{i},P_{j}}\right)$ of the user $U_{i}$ to the provider $P_{j}$ , which is shown in Table 7. The satisfaction information amount $I(P_{j},U_{i})$ of the provider $P_{j}$ to the user $U_{i}$ is shown in Table 8.

Table 7

The amount of information about logistics service users’ satisfaction with providers

	$P_{1}$	$P_{2}$	$P_{3}$	$P_{4}$	$P_{5}$	$P_{6}$	$P_{7}$	$P_{8}$	$P_{9}$	$P_{10}$
$U_{1}$	3.048	7.689	7.934	3.967	2.644	8.907	2.644	9.215	8.760	2.878
$U_{2}$	2.035	8.431	9.426	4.386	1.028	4.139	1.028	9.513	6.897	3.478
$U_{3}$	5.106	3.997	2.894	3.158	4.345	9.201	4.345	2.915	2.785	7.856
$U_{4}$	4.717	7.764	5.770	2.938	9.052	10.865	9.052	1.538	3.482	2.272
$U_{5}$	2.226	1.865	9.208	3.997	1.231	3.491	1.231	8.865	7.248	4.889
$U_{6}$	2.800	3.692	10.352	6.141	4.039	7.552	4.039	1.997	6.741	4.113
$U_{7}$	6.449	4.940	9.082	1.394	7.036	8.296	5.225	0.829	4.957	1.522
$U_{8}$	4.316	7.616	6.294	2.849	3.984	9.443	4.261	2.963	4.292	2.386
$U_{9}$	2.964	2.502	0.214	2.117	4.329	3.212	5.843	0.304	5.137	4.502
$U_{10}$	5.250	3.556	1.612	0.284	5.703	1.394	2.849	0.432	1.038	1.157

Table 8

The amount of information about logistics service providers’ satisfaction with users

	$U_{1}$	$U_{2}$	$U_{3}$	$U_{4}$	$U_{5}$	$U_{6}$	$U_{7}$	$U_{8}$	$U_{9}$	$U_{10}$
$P_{1}$	3.849	0.721	2.885	1.288	0.998	0.883	9.277	7.927	2.495	4.817
$P_{2}$	11.503	10.387	9.592	12.196	1.725	7.896	5.106	1.766	2.784	1.117
$P_{3}$	11.285	7.717	7.312	5.613	11.896	7.083	6.756	2.533	2.618	8.354
$P_{4}$	4.548	12.292	4.363	5.581	4.416	10.029	4.534	9.176	5.852	4.806
$P_{5}$	3.018	11.292	2.617	4.320	3.849	8.955	7.243	6.095	4.066	3.032
$P_{6}$	11.212	8.713	10.943	2.360	11.563	3.887	8.232	3.623	3.022	9.087
$P_{7}$	3.418	1.292	4.414	10.054	7.314	5.613	8.724	7.177	7.619	4.397
$P_{8}$	1.610	3.114	5.004	1.003	0.378	2.332	2.165	3.192	2.017	1.746
$P_{9}$	2.703	0.306	1.935	5.076	3.062	6.236	6.399	2.741	7.417	7.686
$P_{10}$	8.675	5.259	5.944	4.483	8.365	3.111	6.478	3.176	3.236	2.394

(3)

Matching optimal calculation

The multi-objective optimization model of transaction matching is built. In this example, take one-to-one matching as an example, let $\theta_{j}=1$ and $\varphi_{i}=1$ , that is, a logistics service user and a provider trade matching. The principle of one-to-many and many-to-many matching is similar to that of one-to-one matching, where only the values of $\theta_{j}$ and $\varphi_{i}$ need to be changed.

$\displaystyle\text{max}Z_{1}=\sum^{m}_{i=1}\sum^{n}_{j=1}\left({10.865-I\left(% {U_{i},P_{j}}\right)}\right)x_{ij}$ (21) $\displaystyle\text{max}Z_{2}=\sum^{n}_{j=1}\sum^{m}_{i=1}\left({12.292-I\left(% {P_{j},U_{i}}\right)}\right)x_{ij}$ (22) $\displaystyle s.t.\sum^{m}_{i=1}x_{ij}\leqslant 1,j=1,2,\ldots,n.$ (23) $\displaystyle\sum^{n}_{j=1}x_{ij}\leqslant 1,i=1,2,\ldots,m$ (24) $\displaystyle x_{ij}=\left\{{0,1}\right\},i=1,2,\ldots,m,j=1,2,\ldots,n$ (25)

Set the number of chromosomes as 100, crossover probability as 0.9, mutation probability as 0.1, and the maximum number of iterations as NC $=$ 200. The parallelization algorithm NSGA-II based on Hadoop is used, and the solution results are shown in Table 9.

Table 9

Matching optimization solution result

Logistics service transaction matching scheme based on blockchain	Objective function value
	max $Z_{1}$	max $Z_{2}$
$\left\{{{\begin{array}[]{c}{\left({U_{1},P_{5}}\right),\left({U_{2},P_{7}}% \right),\left({U_{3},P_{9}}\right),\left({U_{4},P_{8}}\right),\left({U_{5},P_{% 2}}\right),\left({U_{6},P_{1}}\right),}\\ {\left({U_{7},P_{4}}\right),\left({U_{8},P_{10}}\right),\left({U_{9},P_{3}}% \right),\left({U_{10},P_{6}}\right)}\\ \end{array}}}\right\}$	57.139	54.678

It concludes that $\max Z_{1}=$ 57.139, $\max Z_{2}=$ 54.8, the matching condition is: logistics service users $U_{1}$ trade with providers $P_{5}$ , $U_{2}$ trade with $P_{7}$ , $U_{3}$ trade with $P_{9}$ , $U_{4}$ trade with $P_{8}$ , $U_{5}$ trade with $P_{2}$ , $U_{6}$ trade with $P_{1},U_{7}$ trade with $P_{4}$ , $U_{8}$ trade with $P_{10}$ , $U_{9}$ trade with $P_{3}$ , $P_{6}$ trade with $U_{10}$ .

For matching optimization results, first of all, from the perspective of logistics service users, according to Table 7, the information value of the transaction between user $U_{1}$ and provider $P_{5}$ is the minimum value (2.644), the information value of the transaction between $U_{2}$ and $P_{7}$ is the minimum value (1.028), the information value of the transaction between $U_{3}$ and $P_{9}$ is the minimum value (2.785), the information value of the transaction between $U_{4}$ and $P_{8}$ is the minimum value (1.538), the information value of the transaction between $U_{5}$ and $P_{2}$ is quite small (1.865, ranked third, close to the optimal), the information value of the transaction between $U_{6}$ and $P_{1}$ is quite small (2.800, ranked second, close to the optimal), the information values of transaction between $U_{8}$ and $U_{10}$ is the minimum (2.386), the information values of transaction between $U_{9}$ and $P_{3}$ is the minimum (0.214), the information values of transaction between $U_{10}$ and $P_{6}$ is quite small (1.394, ranked the fifth, close to the optimal). Similarly, from the perspective of logistics service providers, the information value of each pair is relatively small. It shows that the matching results are satisfactory and the algorithm is more effective.

5. Conclusion

At present, there are many problems in the transaction of logistics public information platform, such as the lack of trust, automation and intelligence, etc. With the rapid development of block chain, artificial intelligence and big data technology, the research on intelligent logistics service transaction is gradually emerging. Based on the block chain background, this paper studies the matching of logistics service transaction users and providers, and establishes a multi-objective optimization model of logistics service transaction matching under the block chain through the data standardization of different categories of indicators and the calculation of satisfaction information.

Through block chain logistics service trade matching multi-objective model under the NSGA-II crossover, mutation, genetic inheritance, non-dominated sorting and crowded degree calculation, the NSGA-II algorithm based on Hadoop parallel design. Finally, through the case analysis to get the satisfactory matching results. It has shown that based on Hadoop blocks under the chain of logistics service trade matching the NSGA-II parallel algorithm is feasible.

The following aspects need to be promoted and improved: (1) If more providers or consumers join in the transaction matching, the transaction matching optimization model needs to be improved; (2) Multi-objective matching model of logistics service transaction matching under block chain can be considered to be solved by multi-stage decision.

References

Podgorelec

Turkanovic

and Karakatic

, A machine learning-based method for automated blockchain transaction signing including personalized anomaly detection, Sensors 20(1) (2019), 147.

Shen

X.Y.

and Yuan

, Selling secondhand products through an online platform with blockchain, Transportation Research Part E 142 (2020), 102063.

Niu

B.Z.

Z.H.

Gao

and Gao

, Should multinational firms implement blockchain to provide quality verification? Transportation Research Part E 145 (2021), 102121.

Tang

C.S.

and Veelenturf

L.P.

, The stategic role of logistics in the industry 4.0 era, Transport Res Part E 129 (2019), 1–11.

Yang

C.S.

, Maritime shipping digitalization Blockchain-based technology applications, future improvements, and intention to use, Transport Res Part E 131 (2019), 108–117.

López

and Farooq

, A multi-layered blockchain framework for smart mobility data-markets, Transport Res Part C 111 (2020), 588–615.

Orji

I.J.

Kusi-Sarpong

Huang

and Vazquez-Brust

, Evaluating the factors that influence blockchain adoption in the freight logistics industry, Transportation Research Part E 141 (2020), 102025.

Wang

J.S.

Nie

E.B.

and Cui

J.X.

, Research on credit rating of weak credit group based on block chain technology, Advances in Economics, Business and Management Research 9 (2017), 839–842.

Christidis

and Devetsikioti

, Blockchains and smart contracts for the Internet of Things, IEEE Access 4 (2016), 2292–2303.

10.

Cheung

K.F.

Bell

M.G.H.

and Bhattacharjya

, Cybersecurity in logistics and supply chain management: An overview and future research directions, Transportation Research Part E 146 (2021), 102217.

11.

Yang

M.S.

Jia

Z.J.

Gong

and Zhang

X.F.

, An association ring signature for block chain e-money transactions, Wuhan University Journal of Natural Sciences 24(2) (2019), 169–175.

12.

El-Borhamy

and Mosalam

, On the existence of periodic solution and the transition to chaos of Rayleigh-Duffing equation with application of gyro dynamic, Applied Mathematics and Nonlinear Sciences 5(1) (2020), 93–107.

13.

Shen

and Huang

G.Q.

, Blockchain-enabled workflow operating system for logistics resources sharing in E-commerce logistics real estate service, Computers & Industrial Engineering 135(2019), 950–969.

14.

Kshetri

, Blockchain’s roles in strengthening cybersecurity and protecting privacy, Telecommunications Policy 41(10) (2017), 1027–1038.

15.

Helo

and Hao

, Blockchains in operations and supply chains: A model and reference implementation, Comput Ind Eng 136 (2019), 242–251.

16.

Yadav

and Singh

S.P.

, Blockchain critical success factors for sustainable supply chain, Resources, Conservation & Recycling 152 (2020), 104505.

17.

Choi

T.M.

and Luo

S.Y.

, Data quality challenges for sustainable fashion supply chain operations in emerging markets: Roles of blockchain, government sponsors and environment taxes, Transport Res Part E 131 (2019), 139–152.

18.

Choi

T.M.

Wen

Sun

X.T.

and Chung

S.H.

, The mean-variance approach for global supply chain risk analysis with air logistics in the blockchain technology era, Transport Res Part E 127 (2019), 178–191.

19.

Morkunas

V.J.

Paschen

and Boon

, How blockchain technologies impact your business model, Bus Horiz. 62(3) (2019), 295–306.

20.

Issaoui

Khiat

Bahnasse

and Ouajji

, Smart logistics: Study of the application of blockchain technology, Procedia Computer Science 160 (2019), 266–271.