Construction on Evaluation Index System of Traceability System of Agricultural Internet of Things Based on Network Analysis

Abstract

In order to ensure the transparency and openness of each link of agricultural products “from farmland to table” and to solve the problem of consumers’ low trust degree in the safety of agricultural products and food, the application and evaluation of agricultural Internet of Things traceability system are the top priority. However, the traceability system evaluation indicators are rarely discussed in the existing studies. Consequently, this paper attempts to construct the evaluation index system of the Internet of Things traceability system of agricultural products based on an overall perspective of ecological civilization and to determine the weight of each evaluation index by using analytic hierarchy process (AHP). Then, on the basis of the relevant data of the agricultural product traceability system in recent years, the determined weight is modified by using entropy method. Finally, it points out the important practical value for management departments, production subjects, sales subjects, and consumers, providing reference for improving the quality and safety of agricultural products and standardizing the market supervision of agricultural product market.

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

In recent years, the increasing quality and safety problems of agricultural products have become one of the focus issues under the current social attention and affect the healthy development of China’s agriculture. These horrifying security events include “Sultan Red” salted duck eggs, melamine milk powder, talc flour, artificial red dates to water injection meat, potion bean sprouts, dyed green beans, clenbuterol, sulfur-smoked “poisonous ginger,” and street imitation meat, all of which not only cause great damage to physiology and mental health of customers, but also bring serious impact for agricultural product production enterprises and the whole agricultural product industry chain, restricting the pace of Chinese agricultural enterprises to participate in the international market competition. Agriculture, as China’s primary industry, is facing the challenge of green development. In recent years, the traceability system of the Internet of Things of agricultural products is evaluated scientifically and reasonably with certain sticky problems in the production, processing, and sales of agricultural products improved in accordance with the scientific evaluation results, which is of great significance to ensure the quality and safety of the supply of agricultural products, to promote the green, healthy, and sustainable development of rural agriculture, and to solve the problem of “three rural.”

However, there exist few studies concentrating specifically on evaluating the traceability system of the Internet of Things for agricultural products [1–3]. Especially, due to the fact that the application of the Internet of Things technology in the agricultural field is still in its infancy, most of relevant research mainly focuses on the system design and construction [4–6], and few studies have been conducted to construct the evaluation system of agricultural product Internet of Things traceability system from the overall perspective of ecological civilization [7, 8]. It should be pointed out that the application of the traceability system of the Internet of Things for agricultural products needs to be evaluated to improve each production link of the agricultural product supply chain accordingly, so as to improve the production quality of agricultural products [9]. In addition, in terms of the Internet of Things traceability system of agricultural products, its evaluation index system should not only reflect the overall function of the industrial chain, but also be distinguished between different types of agricultural products.

In view of this, this paper constructs the evaluation index of the traceability system with the perspective of ecological civilization based on reviewing the existing literature about traceability, traceability system, and the evaluation indicators of agricultural product Internet of Things traceability system. And it is evaluated and analyzed by analytic hierarchy process (AHP) and entropy method. Finally, the results and the theoretical and practical contributions of this paper are discussed. In doing so, we make two theoretical contributions: first, a more comprehensive evaluation index system of agricultural product Internet of Things traceability system in line with the development background of the time has been constructed from the perspective of ecological civilization, social, economic, and ecological benefits. Second, we combined the qualitative and quantitative to make the construction of the index system more objective and reasonable. Our research also has important contributions to the safety management of agricultural products: (1) guide the government and other administrative departments to do a good job in supervision, testing, certification, technical support, policy guidance, and financial support; (2) guide the production subjects to do a good job in planting, green and scientific planting, and breeding work; (3) guide the sales subjects to do a good job in the brand, packaging, and sales of agricultural products; and (4) help consumers to understand the traceability system, establish green consumption awareness, and finally form a sustainable cycle of traceability.

2. Related Theory

2.1. Traceability and Traceability System

Regarding the connotation of traceability, the relevant literature mainly defines traceability starting from the concept of traceability. They think that traceability refers to recording digital information on all aspects of the stored product from raw materials or parts to the finished product in order to realize the exchange and tracking of relevant product information between enterprises at each node in the supply chain [10–12]. European Commission defines food traceability as follows: food traceability refers to the ability to trace food, feed, edible livestock, and all substances that may become components of food or feed from the links of source, production, processing, transportation, and sales of food raw materials [13–15]. In terms of the definition of agricultural product traceability system, a representative view holds that it refers to the electronic management of agricultural product production records by recording and storing all kinds of relevant information in the whole process of production, circulation, processing, and sales, so as to establish “identity file” for agricultural products; buyers and consumers can quickly query product information through the identification code, thus achieving the effect of root traceability. The system can provide the breadth of product information, determine the root cause of the problem or the accuracy of product characteristics to the greatest extent, which realizes the seamless connection “from farmland to table,” the integrated dynamic real-time management, and ensure that consumers can buy assured agricultural products, so as to be conducive to the quality management and control of agricultural products [16–18].

In terms of the application of the Internet of Things traceability system for agricultural products, Li et al. studied and developed a flexible and open vegetable quality and safety traceability system suitable for agriculture-related enterprises and professional cooperatives by integrating agricultural Internet of Things technology and key technologies and equipment for quality and safety traceability [19]. Song and Yuan proposed that the agricultural traceability management system stores all links of the crop circulation chain through barcode encryption technology and displays the relevant traceability information to users in the form of map display, which is conducive to realizing early warning and traceability of the crop circulation link [20]. Taking kiwi fruit supply chain as an example, Rahul et al. analyzed the impact of RF identification (RFID) tags on traceability and noted that the traceability system can help track the quality, perishability, and freshness of fruit [21]. Ganjar et al. proposed a pedigree tracing system, which applies radio frequency identification technology to accurately locate agricultural products and uses wireless sensing technology to record the temperature and humidity of the storage and transportation of agricultural product [22].

2.2. Evaluation Index of the Internet of Things Traceability System of Agricultural Products

As a branch of the traceability system, the extensive application of Internet of Things traceability system for agricultural products improves the traceability efficiency of problematic agricultural products and further ensures that the quality of agricultural products is truly effectively supervised and becomes trusted by consumers. Therefore, many scholars, from the perspective of evaluation index, have analyzed the key factors affecting the traceability system through the construction of the primary evaluation index and the secondary evaluation index of the traceability system of agricultural products and provide theoretical significance for the traceability of agricultural products. For example, Chen Song et al. discussed the traceability system of the Internet of Things of agricultural products according to empirical studies, including basic conditions, external factors, internal factors, and standardization degree [23]. Zhou et al. measured the net benefits of the traceability system from the two first-level indicators of income and cost based on the DEA method, so as to measure the final results reflecting the implementation of pork quality and safety traceability by enterprises [24]. Qian et al. used the model to evaluate the traceability granularity of the traceability system of the Internet of Things for agricultural products from three main factors consisting of traceability precision, traceability width, and traceability depth [25]. He et al. evaluated the traceability of aquatic product production and processing enterprises by determining the weights of 3 first-level and 21 second-level indicators [26].

As per the existing literature review, few scholars have considered the important indicators such as green production and the application of the Internet of Things technology in the analysis and evaluation of the Internet of Things traceability system of agricultural products. Therefore, on the basis of the existing research, this paper combines with the formation of national advocacy green development, the production mode of resource conservation, and environmental protection requirements. From the perspective of ecological civilization, it builds an evaluation index system centering on the economic benefits, social benefits, ecological benefits, technical support, and quality and safety guarantee.

3. Construction of the Evaluation Index System for the Traceability System of Agricultural Internet of Things

3.1. Principles of Evaluation Index System Construction

The evaluation index system of the agricultural Internet of Things traceability system is an objective evaluation and reflection of the agricultural product traceability system based on the Internet of Things technology. However, the process of index selection often faces the problem of index redundancy or lack of key indicators. Therefore, the construction of the evaluation index system not only is necessary to be determined according to the main influencing factors affecting the application and development of the agricultural Internet of Things traceability system, but also needs to follow some scientific rules. Specifically, the construction of the agricultural Internet of Things traceability evaluation index system should adhere to the following principles:

(1) Principles of Comprehensiveness and Systematization. On the one hand, the constructed evaluation index should fully reflect the main influencing factors affecting the traceability system. On the other hand, there should be a certain logical relationship among all indicators, so as to evaluate the traceability system systematically and interlinked.

(2) Principles of Minimization and Continuity. There are many types of indicators that reflect the agricultural Internet of Things traceability system; however, there is a substitutability among the listed indicators according to the principle of sufficiency. Therefore, indicators should be screened to build a minimum complete indicator set that can meet the needs. Therefore, the indicators should be screened to build a minimum complete indicator set that can meet the needs. The development of the traceability system of the agricultural Internet of Things is a dynamic process, but the selection of indicators should have relative stability to ensure the indicator system has a certain continuity in time.

(3) Qualitative Principles and Quantitative Principles. Quantitative analysis of the metrics can be realized by evaluating the weight of each index. At the same time, qualitative indicators and quantitative indicators should be combined to objectively evaluate the traceability system of the agricultural Internet of Things for the sake of avoiding some problems that some factors are difficult to quantify.

3.2. Construction of the Evaluation Index System for the Traceability System of the Internet of Things of Agricultural Products

On the basis of fully drawing lessons from the relevant research results at home and abroad [27–30], we summarize and sort out the relevant evaluation indicators given by domestic and foreign documents from five aspects: economic benefits, social benefits, ecological benefits, technical guarantee, and quality and safety guarantee, so as to form the alternative evaluation index for the traceability of the Internet of Things system of agricultural products. After that, based on the definition of agricultural product Internet of Things traceability system, secondary screening of alternative evaluation indicators was carried out by consulting relevant experts in the field of agriculture and actual investigation, which follow the principle of evaluation indicator construction mentioned above. Finally, five first-level indicators and nineteen secondary evaluation indicators were identified. Details are shown in Table 1.

Table 1

Evaluation index system of the Internet of Things traceability system for agricultural products.

Total target layer	Level 1 evaluation index layer	Level 2 evaluation index layer	Index description
Evaluation of traceability system of agricultural products $A$	Economic benefits $B1$	Brand awareness $C11$	It reflects the influence scope or breadth of traceable agricultural product brands
		Market share $C12$ (%)	It reflects the market competitive status and profitability of traceable agricultural products
		Cost return rate of Internet of Things technology application in each link of supply chain $C13$ (%)	The ratio of profit and cost incurred in the use of the Internet of Things technology and reflect the profits derived from unit costs
		Consumer satisfaction $C14$	It reflects consumers’ awareness and willingness to use the agricultural product traceability system
		Government subsidy input $C15$	It affects the input of stakeholders in traceability system, thus affecting the performance of traceability system of agricultural products
	Social benefit $B2$	The authenticity of the information of traceable agricultural products $C21$	It directly affects consumers’ willingness to buy the traceability products and their trust in the traceability system
		Rapid response ability of problem agricultural products $C22$	It reflects the accurate, timely, and rapid traceability ability of unqualified agricultural products, in order to reduce the adverse impact on consumers
		Government supervision system and information disclosure in agricultural products $C23$	It reflects the government’s implementation ability to supervise the traceability system of the Internet of Things for agricultural products and maintains the market order and the interests of consumers in a certain extent
	Ecological benefit $B3$	Producer’s consciousness of green production $C31$	It reflects the degree of environmental pollution that farmers produce agricultural products to some extent
		The used amount of chemical fertilizer and pesticide $C32$	Excessive use of chemical fertilizers and pesticides will cause pollution to the soil, water sources, and air
		Development and introduction rate of environmental protection technology and equipment $C33$	It can promote the green production and processing of agricultural products
		Consumer consumption behaviors of green agricultural products $C34$	It stimulates the production and sales of green produce to some extent
	Technical support $B4$	Home computer and broadband penetration rate $C41$	As the basis of the Internet of Things, the Internet directly affects the monitoring of agricultural production environment and the timeliness of production information transmission to the upper flow
		Mobile phones and the mobile Internet penetration rate $C42$	The use of mobile equipment facilitates the farmers’ management of the production information of agricultural products
		Internet of Things perception technology utilization $C43$	It reflects the degree of intelligence of agricultural product production link; the higher the use degree, the more perfect the “electronic file” of agricultural products
		Agricultural product security information tracking technique level $C44$	The higher the tracking technology, the easier it is for the supervision of agricultural products
	Quality and safety guarantee $B5$	Additive usage $C51$	The use of additives with more than the national standard dose seriously affects the quality and safety of agricultural products
		Packaging material detection $C52$	It ensures the quality and safety of agricultural product processing links
		The certification of “Sanpinyibiao” of traceable agricultural products $C53$	It affects the production and sales of green, organic, pollution-free, and traceable agricultural products

3.2.1. Economic Benefits of the Traceability System of the Internet of Things for Agricultural Products

The economic benefit of the Internet of Things traceability system for agricultural products is a key factor to promote the coordinated development of each link of the agricultural product supply chain. In the process of development, the agricultural product traceability system not only meets the needs of product safety and quality, but also meets the needs of competition and market architecture. As market entities pursuing profits, stakeholders will establish a traceability system and implement it, which will inevitably increase the operating costs of related enterprises. On the one hand, enterprises are only willing to establish traceability systems and implement traceability when the benefit of implementing traceability is no less than the increased costs. On the other hand, brand awareness can not only help sellers open up the market, but also gain consumer trust. In addition, the economic support of governments is the guarantee for promoting traceability products to enter the market [31]. Therefore, considering economic benefit output ratio in the traceability system of investment and use of direct impact, in terms of economic efficiency, this paper evaluates traceable agricultural products from five aspects, including brand awareness, market share, cost return rate of Internet of Things technology application in each link of supply chain, consumer satisfaction, and government subsidized input.

3.2.2. Social Benefits of the Traceability System of the Internet of Things for Agricultural Products

The information disclosure level of the Internet of Things traceability system for agricultural products reflects the degree of information asymmetry in each link of the supply chain, whose authenticity greatly affects the credibility of the traceability system industry. Meanwhile, the negative external spillover problem caused by unqualified agricultural products is bound to have a serious negative impact on the relevant industries. Therefore, the relevant departments of agricultural products need to reduce the incidence of quality and safety accidents of agricultural products through effective supervision means, so as to improve the social benefits of the Internet of Things traceability system of agricultural products. To this end, this paper will analyze the influencing factors of the social benefits of the traceability system of agricultural products from the authenticity of the information of traceable agricultural products, rapid response ability of problem agricultural products, and government supervision system and information disclosure regarding agricultural products.

3.2.3. Ecological Benefits of the Traceability System of the Internet of Things for Agricultural Products

In the problems faced by the traceability of the Internet of Things system for agricultural products, ecological benefits are related to the “vitality” of the Internet of Things traceability system of agricultural products. The environmental-friendly production, sales, and use are the basis of the sustainable development and application of the agricultural Internet of Things traceability system. Agricultural product traceability system of products and origin environment monitoring data acquisition mainly comes from agricultural product planting, processing, storage, transportation, fertilization, weeding, medicine, intervention data and the quality of agricultural product quality grade, pests, and pesticide residue data for comparative analysis [32]. In addition, the main body of the traceability system of agricultural products mainly includes the production, processing, and sales enterprises and consumers of agricultural products. Only with the coordination and cooperation of all parties can the ecological benefits of the traceability system be improved. Therefore, the evaluation of ecological benefit mainly evaluates from four secondary indicators: producer’s consciousness of green production, the used amount of chemical fertilizer and pesticide, development and introduction rate of environmental protection technology and equipment, and consumer consumption behaviors of green agricultural products.

3.2.4. Technical Support of the Internet of Things Traceability System for Agricultural Products

The popularization and use of the agricultural Internet of Things traceability system cannot be separated from the technical guarantee of the Internet, sensing equipment, mobile communication, and other hardware equipment. Internet of Things technology can improve the intelligence degree of production, management, trading, logistics, and other links of agriculture, and promote the development of China’s modern traceability product traceability system. In order to accurately identify traceability data, QR code, tracking code, and product recognition technology play a key role. Therefore, computer technology and Internet of Things perception technology are the technical basis to realize data recognition. But it is not enough to ensure the integrity of agricultural information only has the QR code of each link, it also needs to be tracked. Product information ultimately needs to be passed to the hands of consumers. So, they can use the mobile phone “scan” to obtain the relevant product information to be helped to buy reliable products [33, 34]. Thus, the evaluation of the technical support of the Internet of Things traceability system of agricultural products is mainly designed with indicators from home computer and broadband penetration rate, mobile phones and the mobile Internet penetration rate, Internet of Things perception technology utilization, and agricultural product security information tracking technique level.

3.2.5. Quality and Safety Guarantee of the Traceability System of the Internet of Things for Agricultural Products

Green and environmental-friendly production and processing of agricultural products will not only affect the economic benefits and social responsibilities of the production and processing enterprises, but also affect the nutritional safety of consumers and their willingness to consume the product. So, it is necessary to consider the certification of organic, green, and pollution-free products in order to ensure the safety and green production and processing of traceable agricultural products. In the processing process, enterprises must use additives in strict accordance with the Standards for the Use of Food Additives promulgated by the National Health and Family Planning Commission. At the same time, in the packaging of agricultural products, strict control of testing and packaging materials is also the key elements affecting the quality and safety of agricultural products. Therefore, this paper divides the secondary indicators of the quality and safety guarantee of the traceability system of agricultural products into additive usage, packaging material detection, and the certification of “Sanpinyibiao” of traceable agricultural products.

4. Determining the Index Weight

4.1. Determining the Index Weight by Analytic Hierarchy Process

Analytic hierarchy process is a multi-criterion decision approach combining qualitative and quantitative analysis proposed by Saaty, a famous American operational research scientist, in the 1970s. The basic idea is to decompose a complex problem into an orderly hierarchical structure, namely, target layer, criterion layer, and scheme layer, so as to construct the pairwise comparison matrix, and finally calculate the weight value of each index. This method makes people's thinking process hierarchical and quantitative while providing quantitative basis for analysis, decision-making, prediction, or control with mathematics. The specific steps are as follows.

4.1.1. Establishing a Hierarchical Hierarchy

Through the analysis of complex problems, the first step is to clarify the goal of the decision and take the goal as the element of the target layer (the highest layer), and this target requirement is unique, namely, that the target layer has only one element. Secondly, find out the criteria, affecting the realization of the target, as the criterion layer factor under the target layer. It should be noted that some of the criterion layer factors are the main criteria, and some are sub-criteria which subordinates to the main guidelines. Therefore, criterion elements need to be divided into different levels and groups according to these relationships, and there is general subordination between different levels.

According to the definition and characteristics of the agricultural Internet of Things traceability system, in this paper, the target layer is set as the “evaluation of the agricultural Internet of Things traceability system.” In order to achieve this goal, there are five main criteria to be considered including economic benefits, social benefits, ecological benefits, technical support, and quality and safety guarantee. At the same time, there are 19 sub-guidelines under the main guidelines, so as to form a hierarchical model for traceability system evaluation of the Internet of Things for agricultural products.

4.1.2. Building a Judgment Matrix for Pairwise Comparisons

According to the 1–9 scale method proposed by Saaty, the importance of two metrics under a single criterion is compared and assigned to form a pairwise comparison judgment matrix under a certain criterion. Specific scaling values and meanings are shown in Table 2.

Table 2

1–9-level scaling values.

Importance scaling	Contain
1	It indicates that the two elements compare to each other and they have equal importance
3	It indicates that the two elements compare to each other and the former is slightly more important than the latter
5	It indicates that the two elements compare to each other and the former is significantly more important than the latter
7	It indicates that the two elements compare to each other and the former is much more important than the latter
9	It indicates that the two elements compare to each other and the former is extremely important than the latter
2, 4, 6, 8	Represents the median of the above judgment
Reciprocal value	If the importance ratio of element i to element j is $aij$, the importance ratio of element j to element i is $aji=1/aij$

In order to scientifically and reasonably determine the weight of various indicators of the traceability system of the Internet of Things of agricultural products, this article invited and consulted 15 experts who are in the field of production, processing, and sales of agricultural products, according to the scale value and significance of level 1–9, to compare the significance of 1–9 indicators. After multiple rounds of information interaction, experts build a comparison and judgment matrix of criterion and index layers. Now the criterion layer pairwise judgment matrix shown in Table 3 is taken as an example to assign the weight of the index.

Table 3

Comparative judgment matrix of the criteria $Bi$ under the target $A$.

$A$	$B1$	$B2$	$B3$	$B4$	$B5$
$B1$	1	1	1/2	1/2	1/4
$B2$	1	1	1	1/3	1/2
$B3$	2	1	1	2	1
$B4$	2	3	1/2	1	1/3
$B5$	4	2	1	3	1

4.1.3. Calculating the Weight Vector and Make a Hierarchical Single Ranking

(1) Calculate normalized indicator weight. $Wi={{\displaystyle {\prod }_{j=1}^{n}aij}}^{1/n}/{\displaystyle {\sum }_{i=1}^n{{\displaystyle {\prod }_{j=1}^{n}aij}}^{1/n}},\hspace{1em}i=\mathrm{1,2},\cdots ,n$.

  After calculation, the weight of each index of the agricultural Internet of Things traceability system criterion layer is $Wi=0.105,0.128,0.241,0.182,0.345.$

(2) Calculate the maximum eigenvalue of the judgment matrix: $\lambda \max ={\displaystyle {\sum }_{i=1}^n{AW}_i/nWi},\hspace{1em}\mathrm{1,2},\cdots ,n$.

  Then, put the values into the formula and the maximum eigenvalue of the criterion layer judgment matrix is calculated as $\lambda \max =\mathrm{5.226.}$

(3) Calculate the consistency indicators $CI$, $CI=\lambda \max -n/n-1=\mathrm{0.057.}$

(4) Calculate the consistency ratio $CR�$, $CR=CI/RI$ in which $RI$ is the average random consistency index. The specific values are shown in Table 4 [35]. When $CR=CI/RI<0.1,$ it can be considered that the judgment matrix has a satisfactory consistency; otherwise, the judgment matrix needs to be adjusted.

After calculation, $CR=0.050<\mathrm{0.1.}$ Therefore, the criterion layer judgment matrix has a satisfactory consistency, and the weight of $B1,B2,B3,B4,B5$ is $0.105,0.128,0.241,0.182,0.345.$

The index weight of each index layer can be calculated according to the above methods. Specific values are given in Tables 5–9.

Table 4

Mean random consistency indicators.

n	1	2	3	4	5	6	7	8	9	10	11
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49	1.51

Table 5

Comparative judgment matrix of each index under the criterion $B1$.

$B1$	$C11$	$C12$	$C13$	$C14$	$C15$
$C11$	1	5	1/4	1/2	1/2	$\lambda \max =5.342$
$C12$	1/5	1	1/4	1/3	1/2	$CI=0.085$
$C13$	4	4	1	1	3	$CR=0.076$
$C14$	2	3	1	1	4
$C15$	2	2	1/3	1/4	1

Table 6

Comparative judgment matrix of each index under the criterion $B2$.

$B2$	$C21$	$C22$	$C23$
$C21$	1	1	1	$\lambda \max =3.054$
$C22$	1	1	2	$CI=0.027$
$C23$	1	1/2	1	$CR=0.046$

Table 7

Comparative judgment matrix of each index under the criterion $B3$.

$B3$	$C31$	$C32$	$C33$	$C34$
$C31$	1	1	3	1	$\lambda \max =4.215$
$C32$	1	1	1	2	$CI=0.072$
$C33$	1/3	1	1	1	$CR=0.080$
$C34$	1	1/2	1	1

Table 8

Comparative judgment matrix of each index under the criterion $B4$.

$B4$	$C41$	$C42$	$C43$	$C44$
$C41$	1	3	1/2	3	$\lambda \max =4.174$
$C42$	1/3	1	1/4	1/3	$CI=0.058$
$C43$	2	4	1	2	$CR=0.065$
$C44$	1/3	3	1/2	1

Table 9

Comparative judgment matrix of each index under the criterion $B5$.

$B5$	$C51$	$C52$	$C53$
$C51$	1	1	1/5	$\lambda \max =3.029$
$C52$	1	1	1/3	$CI=0.015$
$C53$	5	3	1	$CR=0.025$

4.1.4. Hierarchical Total Ranking

At first, we assume that the weight of m elements of the layer k − 1 which is relative to the total goal has been calculated as $w^{k-1}={w{1}^{k-1},w{2}^{k-1},\dots ,w{m}^{k-1}}^T.$ And the single ranking weight of the n element of layer k for the j elements of the previous layer is $p{j}^k=p1j^k,p2j^k,\dots ,pn{j}^k$ in which the weight of the elements that are not dominated by J is zero. We represent $p^k=p{1}^k,p{2}^k,\dots ,p{n}^k.$ as the ranking of layer k elements to layer k − 1, and the total ranking of the layer k element for the total target is $w^k={w{1}^k,w{2}^k,\dots ,w{n}^k}^T=p^k{w}^{k-1}.$ Similarly, the total ranking results also require consistency tests. When $CR={\displaystyle {\sum }_{i=1}^{n}w{i}^{k-1}\cdot CIi}/{\displaystyle {\sum }_{i=1}^{n}w{i}^{k-1}\cdot RIi}<\mathrm{0.1.}$, the total ranking has a satisfactory consistency.

After calculation, the total ranking consistency index of agricultural Internet of Things traceability system evaluation index is $CR=0.025<0.1$. And the total ranking results for the 19 secondary levels are shown in Table 10.

Table 10

Total ranking of the secondary index weights.

	$B1$	$B2$	$B3$	$B4$	$B5$	Hierarchical total ranking
	0.105	0.128	0.241	0.182	0.345
$C11$	0.131					0.014
$C12$	0.064					0.007
$C13$	0.359					0.038
$C14$	0.313					0.033
$C15$	0.133					0.014
$C21$		0.327				0.041
$C22$		0.413				0.053
$C23$		0.260				0.033
$C31$			0.321			0.076
$C32$			0.290			0.070
$C33$			0.185			0.045
$C34$			0.205			0.050
$C41$				0.310		0.055
$C42$				0.087		0.016
$C43$				0.425		0.077
$C44$				0.179		0.033
$C51$					0.156	0.054
$C52$					0.185	0.064
$C53$					0.659	0.227

4.2. Entropy Method Modifies the Weight

Entropy method is an objective empowerment method, which determines the index weight by data, so as to make up for the deficiency of analytic hierarchy process in index evaluation. The details are as follows: according to the degree of dispersion of each index data, the entropy right of each index is calculated by using the information entropy, so as to get the index weight more objectively. This paper quantified 19 secondary indicators based on consulting experts and field research and interviews. The data were collected from the National Bureau of Statistics, Brick agricultural data, China database, industry research database, and other related data analysis websites to get the original data of 19 rating indicators from 2010 to 2019.

4.2.1. Data Normalization

Given the inconsistent dimensions of each evaluation index, the raw data require normalization. Suppose that m indicators are given: $X_1,X_2,\dots ,X_m.$, and $X_i=x_1,x_2,\dots ,x_n$; after normalizing it, we will get $Y_1,Y_2,\dots ,Y_m.$.

When the evaluation index is positive,$$\begin{array}{}\text{(1)}& Y_{ij}=\frac{x_{ij}-\min x_i}{\max x_i-\min x_i}.\end{array}$$

When the evaluation index is negative,$$\begin{array}{}\text{(2)}& Y_{ij}=\frac{\max x_i-x_{ij}}{\max x_i-\min x_i}.\end{array}$$

Details are shown in Table 11.

Table 11

Standardization of the evaluation index data.

Index	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
C11	0.4000	0.2000	0.2000	0.0000	0.4000	1.0000	0.0000	0.4000	0.0000	0.2000
C12	0.0000	0.1000	0.2000	0.3000	0.4164	0.5283	0.6444	0.9150	0.8720	1.0000
C13	0.5025	0.5000	0.5000	0.5000	0.0000	0.1242	0.3962	0.5950	0.8997	1.0000
C14	0.0000	0.0000	0.0000	0.0000	0.1613	0.3253	0.4866	0.6532	0.8280	1.0000
C15	0.0000	0.1747	0.4473	0.6575	0.7995	0.9082	1.0000	0.8372	0.8529	0.8529
C21	0.0000	0.1047	0.2094	0.3141	0.4188	0.5100	0.6028	0.7106	0.8536	1.0000
C22	0.0000	0.0530	0.1059	0.1589	0.2368	0.3316	0.5392	0.6909	0.9235	1.0000
C23	0.7939	1.0000	0.7934	0.2204	0.0460	0.0420	0.0000	0.0105	0.4892	0.3259
C31	0.4395	0.4395	0.4395	0.4395	1.0000	0.9404	0.3432	0.0000	0.0329	0.3204
C32	0.7041	0.4831	0.2757	0.1654	0.0369	0.0000	0.0645	0.2674	0.6034	1.0000
C33	0.0000	0.1176	0.2353	0.3529	0.4706	0.5882	0.7059	0.8235	0.9412	1.0000
C34	0.3903	0.3903	0.3903	0.3903	1.0000	0.3203	0.0000	0.1041	0.4280	0.4897
C41	0.0000	0.0757	0.1455	0.2056	0.2180	0.3566	0.4525	0.6273	0.8422	1.0000
C42	0.0000	0.0734	0.1514	0.1939	0.2297	0.4123	0.5292	0.6881	0.8703	1.0000
C43	0.0000	0.0000	0.0815	0.1765	0.2606	0.3750	0.5122	0.6799	0.8323	1.0000
C44	0.0000	0.0329	0.0824	0.1400	0.2257	0.3789	0.4662	0.5766	0.7858	1.0000
C51	0.0000	0.0000	0.0000	0.0000	0.0000	0.0456	0.5841	0.7820	0.9844	1.0000
C52	0.0000	0.1702	0.2713	0.3777	0.5798	0.7819	0.9468	1.0000	0.5904	0.8191
C53	0.0000	0.1667	0.3333	0.4000	0.4333	0.4667	0.5000	0.6667	0.8333	1.0000

4.2.2. Calculating the Weight of Each Index

For an evaluation index, entropy is used to determine the discrete of an index. When the value of the information entropy is smaller, the dispersion of an index is greater, so it will have a greater weight. According to the definition of the information entropy, the calculation formula is as follows: $E_j=-1/\ln n{\displaystyle {\sum }_{i=1}^n{p}_{ij}\ln p_{ij}}$, $p_{ij}=Y_{ij}/{{\displaystyle \sum }}_{i=1}^{n}Yij$. If $p_{ij}=0$, then $E_j$ = 0.

And the calculation formula according to the index weight is ${\beta }_j=1-E_j/K-{{\displaystyle \sum }}^{\text{}}{E}_j$, where K refers to the number of indicators; in this paper K = 19. We can get the weight of each indicator. The specific numerical values are shown in Table 12.

Table 12

The entropy weight.

Index	$E_j$	${\beta }_j$
C11	0.767483036	0.076
C12	0.880093825	0.039
C13	0.909555048	0.030
C14	0.719981847	0.092
C15	0.923119982	0.025
C21	0.887443515	0.037
C22	0.827267884	0.056
C23	0.773499264	0.074
C31	0.874871936	0.041
C32	0.82259534	0.058
C33	0.892272634	0.035
C34	0.9009505	0.032
C41	0.847893615	0.050
C42	0.849622421	0.049
C43	0.813248955	0.061
C44	0.821145649	0.058
C51	0.615723945	0.126
C52	0.904683357	0.031
C53	0.909083411	0.030

4.3. Combining the Entropy Method and Analytic Hierarchy Process to Determine the Weight

Entropy method is empowerment based on the degree of dispersion between the index data, which has the characteristics of high accuracy and strong objectivity. However, the proposed method is easy to ignore the importance of the index itself. If the index data change very little or suddenly larger or smaller, there will be limitations. Analytic hierarchy process can deeply analyze and evaluate the relationship between indicators and their own importance after systematic analysis and judgment, but it will also be affected by subjective factors. Therefore, combining the two methods can compensate for the limitations of both, so as to get a more reasonable index evaluation [35, 36]. The combination empowerment formula is $W_j=\sqrt{{\alpha }_j{\beta }_j}/{{\displaystyle \sum }}^{\text{}}\sqrt{{\alpha }_j{\beta }_j}$.

The details are shown in Table 13.

Table 13

The final indicator weight.

Index	${\alpha }_j$	${\beta }_j$	$W_j$
C11	0.014	0.076	0.036
C12	0.007	0.039	0.018
C13	0.038	0.030	0.037
C14	0.033	0.092	0.061
C15	0.014	0.025	0.021
C21	0.041	0.037	0.043
C22	0.053	0.056	0.061
C23	0.033	0.074	0.055
C31	0.076	0.041	0.062
C32	0.070	0.058	0.071
C33	0.045	0.035	0.044
C34	0.050	0.032	0.045
C41	0.055	0.050	0.058
C42	0.016	0.049	0.031
C43	0.077	0.061	0.076
C44	0.033	0.058	0.049
C51	0.054	0.126	0.091
C52	0.064	0.031	0.050
C53	0.227	0.030	0.091

4.4. Results

According to Table 13, the certification of “Sanpinyibiao” of traceable agricultural products (C₅₃), utilization rate of the Internet of Things sensing technology (C₄₃), producer’s awareness of green production (C₃₁), and application amount of chemical fertilizers and pesticides (C₃₂) have a high impact on the evaluation system of Internet of Things traceability system of agricultural products. Therefore, when using Internet of Things traceability system of agricultural products, farmers and other producers should pay more attention to the certification of “Sanpinyibiao” and enhance their own awareness of green production as well as use as much as possible chemical fertilizer below the national standard dosage, so as to ensure the quality and safety of the tea leaves produced and guarantee consumers to consume with the peace of mind. In addition, in the utilization of the traceability system, relevant departments and enterprises should increase the investment of Internet of Things technology, improve the farmers’ cognitive level of the Internet of Things technology by newspapers, news, visits, and other ways, and expand and standardize the size of the RFID market so as to provide strong technical support for the agricultural product Internet of Things traceability system.

5. Discussion

Based on the existing research of the traceability system and the evaluation index of the traceability system of the Internet of Things, combined with the current background of environmental protection and green development, and in view of the shortcomings of the existing research, the evaluation index system of traceability of agricultural Internet of Things from the perspective of ecological civilization is constructed. Later, based on the hierarchical relationship between the indicators, the weight of rating index of the agricultural product Internet of Things traceability system was determined according to the AHP and entropy method. The study results show that on the one hand, based on AHP and entropy method the evaluation index weight of agricultural Internet of Things traceability system can be calculated and evaluated. On the other hand, the traceability focus of different agricultural products has significant differences. For example, the focus of the fruit and vegetable traceability system is the production link and storage and transportation link, but the wild mountain bacteria are the processing link.

The innovation of this paper is that of the following: firstly, based on the new perspective of ecological civilization, the evaluation index system of agricultural product Internet of Things traceability system is constructed with economic benefits, social benefits, ecological benefits, technical support, quality and safety guarantee as the guidelines. Previous studies have constructed the evaluation indexes of different types of traceability systems based on different criteria. However, few studies have comprehensively incorporated economic, social and ecological benefits into the evaluation system to fully reflect the significance of the traceability system for the safety of agricultural products and human health. Secondly, we combined AHP and entropy method to construct the index system with the combination of qualitative and quantitative analysis. The AHP not only integrates the opinions of experts in related fields, but also fully absorbs the research results of the existing literature from a qualitative point of view. The entropy weight complements the shortcomings of qualitative research methods with quantitative methods, so as to make the construction of the index system more scientific and reasonable.

6. Managerial Implications

With the rapid development of the Internet of Things, big data, block chain, and various kinds of information technologies, the whole society will pay more attention to the security of agricultural products. Therefore, the traceability system of the Internet of Things will be more widely used in the whole industry chain of the production, processing, and sales of agricultural products. The establishment of the evaluation index system of the agricultural product Internet of Things traceability system can play different effects in the agricultural product traceability system and provide management guidance for the application of each subject.

Firstly, as for the government and other administrative departments are responsible for the production, processing, transportation, and sales of agricultural products.On the one hand, the evaluation index system can help managers to judge and identify the timeliness of the regulatory system and information disclosure, carry out targeted information tracking and management, and do a good job in all kinds of testing and certification work. On the other hand, through the evaluation of technical support and other indicators, it can not only assist relevant departments to accurately increase the investment of agricultural Internet of Things technology and environmental protection equipment, but also ensure the comprehensiveness and effectiveness of the technology application. At the same time, it can also provide a basis for the government departments to provide the policy and fund support and strengthen the training and guidance to the relevant enterprises.

Secondly, as for the main body of agricultural products, the evaluation index system not only takes into account economic benefits, but also fully considers ecological benefits and social benefits, which can provide producers with how to actively apply the Internet of Things technology from the perspective of ecological civilization to realize the precision and intelligent production and processing, and strengthen the framework system of green production and green processing consciousness. The application of the traceability system can also realize the road of characteristic agriculture from the technical level, provide technical support for scientific planting and breeding, and provide a guarantee for the intensive management and large-scale economy of agricultural products.

Thirdly, as for agricultural products sales subjects, the application of the Internet of Things traceability system and the establishment of the evaluation index system will give full play to the advantages of information. It can guide the sales entities to expand the brand and packaging operation, and actively promote the business coordination and technical docking between the traceability system and the business credit system. What’s more, it can also further use the standard system to guide various consumers to be given priority to the procurement of traceable products, and promote and encourage the cooperation and joint operation of upstream and downstream enterprises.

Finally, as for consumers, the systematic cognition of the agricultural product traceability system has not yet been established, and its evaluation standards are not perfect. The construction of this index system can not only help consumers to establish the awareness of green consumption, but also help them to understand the evaluation standards of the agricultural product traceability system. Consumers’ cognition and acceptance of traceability of agricultural products will continuously promote the safety production awareness of agricultural production subjects and sales subjects, and stimulate their endogenous power of safe production, so as to realize the traceable and sustainable operation of agricultural products and truly realize the whole process guarantee of agricultural products “from the farm to the table.”

Acknowledgments

This work was supported by Yunnan Province Talent Training Project of China (No. 14119073).

References

[1] H. Y. Wu, S. S. Chen, X. Y. Dai, "Study on fuzzy comprehensive evaluation of agricultural products closed-loop supply chain system," Logistics Technology, vol. 45 no. 8, 2022.

[2] G. B. Zhang, G. F. Su, "Construction of agricultural products E-commerce from the spective of ecosystem," Marketing Week, vol. 4, pp. 53-55, 2020.

[3] T. Jia, "Design of the prototype of the logistics evaluation system for fresh agricultural products," Guangdong Silkworm Industry, vol. 52 no. 5, 2018.

[4] L. S. Li, "Application of Internet of Things technology in modern development of agriculture," Chinese edible fungus, vol. 38 no. 5, 2019.

[5] Y. J. Yue, X. F. Yue, Y. Y. Zhong, "Research progress of agricultural Internet of Things system architecture and key technologies," China Agricultural Science and Technology Guide, vol. 21 no. 4, pp. 79-87, 2019.

[6] Y. C. Ou, P. Cao, W. Xu, "Development framework and policy suggestions of agricultural Internet of Things in China based on business ecosystem," Business and Economic Research, vol. 24, pp. 131-134, 2019.

[7] T. Ma, C. H. Sun, W. Y. Li, "Design and implementation of agricultural product origin trusted traceability system based on NB-IoT," China Agricultural Science and Technology Guide, vol. 21 no. 12, pp. 58-67, 2019.

[8] Y. J. Huang, "Research on tea traceability system based on RFID and Internet of Things technology," vol. 16 no. 4, pp. 112-117, 2016.

[9] H. W. Bai, A. D. Sun, J. Chen, "Agricultural product quality and safety traceability system based on the Internet of Things," Jiangsu Agricultural Journal, vol. 29 no. 2, pp. 415-420, 2013.

[10] Z. C. Wu, Z. Y. Xu, "Key technologies of agricultural Internet of Things and its application in agricultural product quality traceability system," China Venture Capital, vol. 10, pp. 41-42, 2010.

[11] Y. P. Liu, F. G. Xie, "Research on the ways to improve the operation efficiency of food traceability system in China," Modern Management Science, vol. 6, pp. 78-80, 2014.

[12] H. L. Li, X. W. Wen, "Research and application of fresh agricultural products supply chain safety," Science and Technology Management Research, vol. 1, pp. 209-212, 2011.

[13] L. Zhang, H. Yu, M. X. Ma, "Information system platform for quality and security of agricultural products based on Internet of Things," Chinese Science, vol. 40, pp. 216-225, 2010.

[14] R. H. Huang, PRODUCT TRACING SYSTEM ANALYSIS AND DESIGN, 2006.

[15] P. Begona, G. Alberto, Food Teaceaility Around the World,” Traceability of Tuna, Iberian Pork and Wine1, pp. 89-103, 2008.

[16] K. Y. Jiang, Y. C. Zhang, X. L. Liao, "Exploration and traceability of the quality and safety of agricultural products based on the Internet of Things technology," Journal of Zhejiang Agricultural Sciences, vol. 8, pp. 1296-1298, 2014.

[17] L. L. Chen, F. J. Yu, "Analysis on the construction of agricultural product quality and safety traceability system in China," Quality and Safety of Agro-Products, vol. 4, pp. 62-65, 2014.

[18] X. T. Yang, J. P. Qian, C. H. Sun, J. T. Zeng, "Key technologies for establishment agricultural products and food quality safety traceability systems," Transactions of the Chinese Society for Agricultural Machinery, vol. 11, pp. 212-222, 2014.

[19] Y. S. Li, C. Wang, Z. Y. Sun, "Design and implementation of the vegetable Quality and Safety traceability system based on the Internet of Things," Information of Agricultural Science and Technology, vol. 14, pp. 35-36, 2016.

[20] L. L. Song, L. Yuan, "Design and implementation of the crop traceability system based on GIS," JIANGXINONGYE, vol. 18, 2018.

[21] R. Gautam, A. Singh, K. Karthik, S. Pandey, F. Scrimgeour, M Tiwari, "Traceability using RFID and its formulation for a kiwifruit supply chain," Computers & Industrial Engineering, vol. 103, pp. 46-58, DOI: 10.1016/j.cie.2016.09.007, 2017.

[22] G. Alfian, J. Rhee, H. Ahn, J. Lee, U. Farooq, M. F. Ijaz, M. A Syaekhoni, "Integration of RFID, wireless sensor networks, and data mining in an e-pedigree food traceability system," Journal of Food Engineering, vol. 212, pp. 65-75, DOI: 10.1016/j.jfoodeng.2017.05.008, 2017.

[23] S. Chen, N. Wang, Y. Z. Qian, "Empirical study of the influencing factors of government departments in promoting the construction of agricultural product quality and safety traceability system," Quality and Safety of Agro-Products, vol. 6, 2013.

[24] J. H. Zhou, K. Li, X. L. Chen, "Study on the strategies of pork traceability system constructing: based on cost-benefit analysis of implementing traceability system in slaughtering and processing section," Issues in Agricultural Economy, vol. 34 no. 10, pp. 90-96, 2013.

[25] J. P. Qian, X. X. Liu, X. T. Yang, "Construction of index system for traceability granularity evaluation of traceability system," Transactions of the Chinese Society of Agricultural Engineering, vol. 30 no. 1, pp. 98-104, 2014.

[26] J. He, W. X. Liu, W. R. Wang, "Establishment of effectiveness evaluation index system of quality safety traceability system of aquatic products manufacturing enterprise," The Food Industry, vol. 12, pp. 260-264, 2018.

[27] Y. S. Liu, "Transparency evaluation index system of pork supply chain based on process," Jiangsu Agricultural Sciences, vol. 46 no. 16, pp. 334-338, 2018.

[28] L. Chen, Research on the Construction of Safety Supervision and Evaluation Index System in Food Sales Link, 2017.

[29] W. Y. Guo, "Study on the construction of the index system for evaluating the performance of food traceability system," Science and Technology Management Research, vol. 37 no. 10, pp. 81-87, 2017.

[30] Y. M. Chiang, W. L. Chen, C. H. Ho, "Application of analytic network process and two-dimensional matrix evaluating decision for design strategy," Computers & Industrial Engineering, vol. 98, pp. 237-245, DOI: 10.1016/j.cie.2016.06.005, 2016.

[31] Y. L. Shi, "The influence of agricultural product traceability system on brand and its enlightenment on marketing," Modern Marketing, vol. 6, pp. 70-71, 2019.

[32] J. Li, M. R. Fu, Y. K. Lu, "Problems and countermeasures of traceability management of agricultural products in China," Chinese fruit vegetables, vol. 41 no. 2, 2021.

[33] Y. Yuan, Y. W. Ou, "Research and implementation of traceability Product quality monitoring of traceability System," Technology Innovation and Application, vol. 23, pp. 106-107, 2020.

[34] Y. Wang, "Design of modern agricultural product traceability system based on the Internet of Things," Digital Technology and Application, vol. 40 no. 7, pp. 184-186, 2022.

[35] S. B. Xu, Practical Decision Method- -Hierarchical Analysis Principle, 1988.

[36] N. Xu, "Research on performance evaluation of aquatic products cold chain transportation based on AHP and entropy weight method," Scientific Journal of Economics and Management Research, vol. 3 no. 8, 2021.