Comprehensive Evaluation and Promotion Strategy of Agricultural Digitalization Level

Abstract

The development of digitalization is a crucial aspect of agricultural progress, and expediting the establishment of digital systems is a significant driving force behind high-quality agricultural advancements in the current era. Utilizing data from 16 cities within Shandong Province in China between 2014 and 2020, we created an assessment system to measure the degree of agricultural digitalization, utilized the entropy technique to assess the level of digitalization, scrutinized the general trends and time-dependent features of each city, and then utilized the obstacle degree model to pinpoint the primary hindrances to digitalization in agriculture. Lastly, the ESDA method was utilized to examine the differences in spatial distribution among regions and the spatial characteristics of agricultural digitalization at different stages and levels. Overall, the degree of agricultural digitalization can be categorized into three stages: deceleration and upswing (2014–2015), steady fluctuation (2016–2017), and high-level upswing (2018–2020). From the perspective of obstacles, the main hurdles to agricultural digitalization are e-commerce transaction volume and the total amount of telecommunication business. To accelerate the development of the entire agricultural industry chain, it is required to leverage the strengths of high-value areas and reinforce the coordination mechanism among various departments while hastening the construction of rural infrastructure in low-value areas. Additionally, it is necessary to improve inter-regional communication and cooperation to nurture different regional development models in line with local conditions.

1. Introduction

In order to vigorously promote the development of agriculture digitalization, realize agricultural modernization and sustainable development. Among them, agricultural digital construction refers to improving the output and efficiency of modern agriculture by making full use of advanced digital technologies and a large number of digital products in our national economy [1]. However, the digital development of Chinese agriculture and rural areas is still in initial exploration. There is still a big gap between the implementation of agricultural and rural modernization and high-quality development, and the digital application level, digital infrastructure construction, and the rural digital development environment. According to “White Paper on the Development of China’s Digital Economy (2021)” [2], the proportion of the agricultural digital economy in industrial added value in China in 2020 is only 8.9%, 0.7 percentage points higher than that in 2019, but the level of agricultural digitalization is still low, far lower than the average level of digitalization in the whole industry. Agricultural digitalization uses blockchain, big data, and cloud platforms to provide data information on precision production, processing, and circulation, which is an inevitable requirement for the green and sustainable development of agriculture; therefore, it is required to pay more attention to agricultural digitalization, accelerate the development process of agricultural digitalization.

Shandong Province is a big agricultural province and a typical area of digitalized rural agriculture development. Since 2016, Shandong Province has been taking new steps toward digitalized agriculture. In 2021, Shandong issued the Plan for the Construction of a Strong Digital Province in the 14th Five-Year Plan of Shandong Province. In developing an integrated and innovative digital economy, Shandong mentioned that it would promote and accelerate the development of featured and efficient digital agriculture. By the end of 2020, the household penetration rate of broadband will reach 96.4%. The online retail sales of agricultural products reached 32.9 billion yuan, up by more than 25% year on year; There are 26 national e-commerce demonstration counties in rural areas, and a diversified e-commerce demonstration system has taken shape [3]. However, there are still some outstanding problems. For example, the construction of agricultural digital infrastructure in Shandong Province is relatively slow, and agricultural Internet of Things equipment and intelligent agricultural machinery equipment have not been widely popularized. In addition, there are also problems such as insufficient deep integration of digital technology and agriculture, and the relative shortage of agricultural digitalized talents, which affect the development process of agricultural digitalization. Therefore, this paper calculates the level of agricultural digitalization and considers the readiness of agricultural digitalization construction in Shandong Province. The purpose of this study is to analyze the effect of agricultural digitalization and propose targeted policies according to the development situation of different regions to improve the level of sustainable agricultural development.

2. Literature Review

The research on agricultural digitalization has been the focus of the academic circle in recent years. China is transforming and upgrading from an agricultural power to an agricultural power. Promoting sustainable agricultural development through digitalization is a sufficient guarantee for building agricultural power. Under the new situation, agricultural digital transformation has gained some practical experience and gains in exploring and developing product-for-sale. Domestic, and foreign scholars mainly focus on the research of agricultural digital development in the following aspects.

2.1. Agricultural Digitalization

The degree of digitalization in agriculture is a key factor in the implementation of agricultural progress. In 2020, Chinese agriculture is in a critical period of transformation and upgrading. Digital means and information technology will run through the whole process of agricultural production development. Recent studies focused on the improvement in energy levels and environmental efficiency, and the reduction of environmental emissions due to climate change [4,5,6,7,8] without considering the application of digital agriculture. The development of digital agriculture has made certain achievements, and the digital transformation of agriculture has begun to take effect. At the same time, promoting the digital transformation of traditional agriculture is a strategic measure to conform to agricultural modernization [9,10], and an inevitable path to promote rural construction and realize agricultural power [11]. Introducing digital innovation and information technology to agricultural efficiency can improve the efficiency of the production process [12]. It is crucial to improve productivity, agricultural production efficiency and agricultural information access [13], which greatly improves the level of digital infrastructure construction and service supply [11]. To realize agricultural modernization and improve agricultural comprehensive benefit and competitiveness, it is required to promote the digital development of high-quality agriculture. The rapid development of agricultural digitalization has created favorable conditions for training and education in digital agriculture, specifically in the fields of robotics, artificial intelligence, and big data analysis [14,15].

2.2. Gap in the Development of Agricultural Digitalization

If societies do not adopt digital agriculture, they risk failing to achieve sustainability and food system security. One of the major concerns of Australian producers with regard to agricultural digitization is the lack of transparency and clarity around issues of data ownership, privacy, trust, and agricultural data security [6,16]. China’s agricultural and rural digitalization transformation is in the preliminary exploration stage, and the application level of agricultural digitalization needs to be further improved. There are shortcomings in infrastructure, application services, and other aspects, and the economic growth point brought by digital technology is still weakly related to the livelihood of most farmers [17]. Although digital transformation has become an important driving force to improve the quality and efficiency of agriculture and realize modernization, and the process is gradually accelerating and the scale is growing, there are still many problems affecting the digital transformation of agriculture. From the perspective of technology and talents, Luo Junwen in 2020 pointed out that the current challenges facing Chinese agricultural digitalization are low technical efficiency of elements and a lack of high-quality talents [18]. The development level of agricultural digitalization is low, the overall construction of complete data resources is lacking, and the digital application of agricultural operating subjects is insufficient, which leads to the bottleneck of digital technology research and development, the conversion rate of scientific and technological achievements among digital agricultural research institutions is low, and the innovation and synergy of related supporting services are insufficient [19,20,21,22]. As well, digital transformation in agricultural practices may lead to a “digital divide” between small and large farms due to the characteristics and availability of digital technology [23]. The above phenomena reflect the problems of digital infrastructure, digital technology, talents, related supporting facilities, and other aspects in the process of agricultural digital development, which directly affect the high-quality development of digital agriculture.

2.3. Strategies for Implementing Agricultural Digitalization

Many countries have adopted a systematic approach to developing agricultural digitalization. Albania, Turkey and other countries have concluded international agreements to participate in the development of national strategies for digital agriculture with agricultural organizations within the framework of providing assistance. The USA and Germany have compared the issues that surround the adoption of digital technology on the farm that will foster more environmentally sustainable food processing systems [24,25]. Also included are sensors to support dairy farm health management during agricultural digitalization [26,27]. The development of agricultural digitalization includes the collection of cross-fields and multiple technologies. It is a complex ecosystem problem featuring multi-business interaction, multi-agent collaboration, multi-mechanism link, age and multi-element coordination [28]. Construct diversified, three-dimensional, and interactive ways and approaches such as collaborative innovation networks based on the industrial chain to promote the smooth implementation of digital transformation [29]. At the same time, it is essential to improve digital facilities and services in rural and poor areas to promote the digital transformation of agriculture and rural areas [11]. Agricultural digital transformation in the new era includes four aspects: digitalization of agricultural production, digitalization of agricultural sales, digitalization of quality and safety of the whole agricultural industry chain, and digitalization of infrastructure [18]. To achieve a high-quality digital transformation of China’s agriculture, the market should play a decisive role in allocating data resources, and the government should play a better role [6]. The continuous development of agricultural digitalization can carry out intelligent and precise control over the whole process of agricultural input, production, and circulation. Agricultural remote sensing and agricultural digitalization should be strengthened [29], and appropriate scale and intensive development should be guided for integrated subjects [30]. It is also necessary to innovate the development concept of agricultural digital transformation, which focuses on the fundamental reform of agricultural digital production and operation through data reconstruction of agricultural production factor allocation efficiency [31].

To sum up, the existing experts and scholars have made active exploration of the construction of agricultural digitalization. There have been sufficient studies on the necessity, existing problems, realization path, and other theoretical contents of agricultural digital transformation. The research on agricultural digitalization hopes to realize agricultural precision production, reduce agricultural production risks and costs, promote sustainable development, and make the agricultural production process more energy-saving and environmental protection. By referring to the previous research ideas of experts and scholars, this paper calculates the development level of agricultural digitalization in Shandong Province, sorts out and discusses the effectiveness of agricultural digitalization transformation in Shandong Province, and based on the current agricultural development stage of Shandong Province, summarizes practice experience, finds problems and difficulties, researches and excavates typical cases, and considers the differences in rural areas of Shandong Province. Different types, local conditions, and different modes of agricultural digital transformation promotion strategies have positive practical guidance for further accelerating agricultural digital transformation according to local conditions.

3. Materials and Methods

3.1. Data Sources and Processing

In this paper, the agricultural digitalization construction level of various cities in Shandong Province during 2014–2020 is taken as the research object, and the data used are mainly from the Statistical Yearbook of Shandong Province (2014–2020), the statistical yearbook of each city and the statistical bulletin. In the selection of indicators, the mean method is used to complete some missing data. As there are certain differences in the dimensions, attributes, and units of the evaluation index system, extreme value method is applied to the evaluation index data to eliminate the impact:

For positive indicators, the original data standardization formula is:

$${X^{\prime }}_{ij}=\frac{X_{ij}-min{X}_j}{max{X}_j-min{X}_j}$$

(1)

For negative indicators, the original data standardization formula is:

$${X^{\prime }}_{ij}=\frac{max{X}_j-X_{ij}}{max{X}_j-min{X}_j}$$

(2)

where $X_{ij}$ represents the original value of the index, ${X^{\prime }}_{ij}$ represents the standardized value, $max{X}_j$ represents the maximum value of index $j,min{X}_j$ represents the minimum value of index $j$.

3.2. Index Selection and System Construction

In the selection of the readiness index, this study draws on the relevant indicators taken from previous studies [32,33,34] and the Evaluation Report on the Development Level of National County Digital Agriculture and Rural Development, and fully combined the research results of current scholars’ index construction system of digital agriculture and digital countryside. Therefore, in accordance with the principles of representativeness, integrity and data availability, this paper selects 16 indicators from four levels: digital agricultural application environment, digitalization of agricultural production, infrastructure and digitalization of industry, and constructs an evaluation index system for the readiness of agricultural digitalization development, as shown in

Table 1. Construction of evaluation index system of agricultural digitalization construction readiness.

First-Level Index	Weights	Second-Level Index	Unit	Weights	Attribute
Digital agricultural application environment	0.2408	C1 Rural electricity consumption	10,000 KWH	0.0499	+
		C2 Average salary of Employees in Information transmission, Software and Information Technology Services	RMB	0.0245	+
		C3 Local financial Expenditure on Science and Technology	Ten thousand RMB	0.0840	+
		C4 Per capita expenditure on general public services	RMB/person	0.0824	+
Digitalization of agricultural production	0.2443	C5 Total power of agricultural machinery	kW	0.0517	+
		C6 Effective irrigated area	thousands of hectares	0.0478	+
		C7 Agricultural plastic film usage per unit area	ton/ha	0.0794	-
		C8 Agricultural diesel oil quantity	ton	0.0654	-
Digital Infrastructure	0.1735	C9 Number of mobile phone users	10,000 households	0.0514	+
		C10 Internet Broadband access Users	ten thousand households	0.0487	+
		C11 Expressway mileage per person	km/person	0.0346	+
		C12 Computers per 100 population	units	0.0388	+
Digitalization of industry	0.342	C13 Total Postal Services	100 million RMB	0.0868	+
		C14 Total telecommunications business	100 million RMB	0.0974	+
		C15 Per capita Gross output value of agriculture, forestry, animal husbandry, and fishery	RMB/person	0.0371	+
		C16 E-commerce Turnover of agricultural products	ten thousand RMB	0.1207	+

.

3.3. Empirical Framework

3.3.1. The Entropy Method

To reduce the deviation of index measure, this study chooses the relatively objective and scientific entropy method for objective weighting. It calculates the information entropy of indicators and determines the weight of indicators according to the influence of the relative change degree of indicators on the whole system, which is a mature weighting method. The specific steps of the entropy method are as follows:

Calculate the proportion of index $j$ in year $i$, then

$$P_{ij}=\frac{{X^{\prime }}_{ij}}{{\sum }_{i=1}^m{X^{\prime }}_{ij}}$$

(3)

$m$ is the number of years participating in the evaluation.

Calculate the index information entropy, then

$$e_j=-k{\sum }_{i=1}^m\left(P_{ij}ln{P}_{ij}\right)$$

(4)

$k=\frac{1}{lnm}$, 0 ≤$e_j$≤ 1.

Calculate the information utility value, then

$$d_j=1-e_j$$

(5)

Calculate the index weight, then

$$w_j=\frac{d_j}{{\sum }_{j=1}^n{d}_j}$$

(6)

After the weight of index $j$ in the system is obtained, the score of index $j$ in year $i$ is obtained

$$df=w{X^{\prime }}_{ij}$$

(7)

Meanwhile, Shandong Province is taken as the study area in this paper. Laiwu City of Shandong Province is not included in the study due to the adjustment of administrative division cancellation.

3.3.2. Obstacle Degree Model

To deeply consider the impact of each subsystem on the development level of agricultural digitalization, it is necessary to introduce the obstacle degree model for diagnosis and analysis. The factor contribution degree, index deviation degree and obstacle degree are used to analyze and diagnose the main obstacles in the development of agricultural digitalization. The following obstacle degree model is constructed:

$$u_j=w_j$$

(8)

where, factor contribution degree $u_j$ represents the impact degree of index $j$ on the overall goal, namely, the weight $w_j$ of a single factor on the overall goal. The index deviation degree $F_{ij}$ represents the secondary index and the agricultural digital development index, that is, the difference between the standardized value ${X^{\prime }}_{ij}$ of a single index calculated by the extreme value method and 100%.

$$F_{ij}=1-{X^{\prime }}_{ij}$$

(9)

Obstacle degree $O_{ij}$ represents the influence value of secondary index on agricultural digitalization development, namely one-way index obstacle degree. The order of this value can determine the primary and secondary relationships of obstacles in the development of agricultural digitalization.

$$O_{ij}=\frac{F_{ij}·w_j}{{\sum }_{j=1}^n{F}_{ij}·w_j}$$

(10)

3.3.3. Exploratory Spatial Data Analysis

ESDA is a collection of spatial data analysis methods and techniques. With spatial correlation as the core, ESDA can find spatial clustering and spatial anomalies. In this paper, the spatial distribution and evolution trend of agricultural digitalization development in Shandong Province were discussed, and the natural breakpoint method in ArcGIS10.8 was used to divide the development level of agricultural digitalization in Shandong Province. The “clustering” method was used to make the internal similarity of each group the greatest, and the differences between the external groups the greatest.

4. Results and Discussion

4.1. Overall Change Trend of Agricultural Digitalization Level

The original data on the agricultural digitalization development level in Shandong Province from 2014 to 2019 were standardized to calculate the agricultural digitalization level. On this basis, the annual growth rate of agricultural digitalization level in Shandong Province was obtained (Figure 1).

The overall level of agricultural digitalization in Shandong Province showed an upward trend, from 2.5431 in 2014 to 4.6112 in 2020, with an average annual growth rate of 10.49% and a “wavy” change. The highest point of growth rate was 16.19% in 2018, and the lowest point was 5.72% in 2016. According to the figure, the development process of agricultural digitalization level in Shandong Province can be divided into three stages: deceleration and rise stage (2014–2015), steady fluctuation stage (2016–2017), and high-level rise (2018–2020). In the first stage, the growth rate decreased and reached its lowest point, but the growth rate was always positive, and the agricultural digitalization level increased slightly. In the second stage, the growth rate is stable, fluctuating around 5%, and the level of agricultural digitalization is rising steadily. After the first two stages of development, the growth rate increased to 16.19% in 2018, and the digitalization level of agriculture has significantly improved. After 2018, the score exceeded 3.5, entering the high-level stage, but the growth rate slowed down compared to 2018, because the country began to pay more attention to quality and no longer simply pursue excessive growth rate to achieve tangible and sustainable growth.

4.2. Overall Change Trend of Agricultural Digitalization Level

The internal structure of the agricultural digitalization level in Shandong Province includes four first-level indicators: digital agricultural application environment, digitalization of agricultural production, digital infrastructure, and industrial digitalization, and the data from 2014–2020 are calculated (Figure 2). In the internal structure, the increased range of the first level index: industry digitalization > agricultural production digitalization > digital agricultural application environment > digital infrastructure. The specific evaluation results are as follows:

Digital agriculture application environment: from 2014 to 2020, the overall fluctuated upward trend, showing a “hump” shape, but the internal structure has the smallest increase. It reached its peak in 2016, with the highest score, which increased from 0.0592 in 2014 to 0.072 in 2016, then decreased to 0.0615 in 2018, and then increased slightly. Further analysis shows that in the secondary index of the digital agricultural application environment, rural electricity consumption played the biggest role and scored the highest during 2014–2017. From 2018 to 2020, local financial expenditure on science and technology made the biggest impact on the application environment of digital agriculture.

Digitalization of agricultural production: From 2014 to 2020, it showed a slightly fluctuating upward trend, but in 2016, there was a significant increase, with an increase of up to 86%, which first decreased from 0.0567 in 2014 to 0.0548, then increased to 0.1018, and finally to 0.0599. Further analysis shows that among the secondary indexes of digitalized agricultural production, the amount of agricultural diesel has the greatest effect on the application environment of digital agriculture in most years. In 2016, the amount of agricultural plastic film per unit area has the greatest impact.

Digital infrastructure: from 2014 to 2020, it showed a slight decline, with a small decline of 9.36% in 2016, and then it rose to 0.0297 in 2017 and then all the way down to 0.0254 at last. Further analysis shows that among the secondary indexes of digital infrastructure, the four secondary indexes have basically the same degree of influence on digital infrastructure and play their roles evenly.

Industrial digitalization: It showed an upward trend from 2014 to 2020. Although there was a decrease in 2016, it still had the largest growth rate among the internal structure, rising from 0.0687 in 2014 to 0.1103, and finally slightly decreased to 0.0905. Further analysis shows that e-commerce sales have been playing an extremely important role in industrial digitalization among the secondary indexes of industrial digitalization.

4.3. Analysis on Time Sequence Characteristics of Agricultural Digitalization

According to Formulas (3)–(7), the agricultural digitalization construction level of cities in Shandong Province from 2014 to 2020 was measured, and the cities were ranked accordingly.

Table 2. Comprehensive scores and rankings of some year readiness evaluation of agricultural digitalization construction in Shandong cities.

Ranking	2014		2016		2018		2020
1	Qingdao City	0.3327	Qingdao City	0.4073	Qingdao City	0.5714	Qingdao City	0.6154
2	Yantai City	0.2314	Yantai City	0.2574	Jinan City	0.3439	Jinan City	0.4797
3	Weifang City	0.2270	Jinan City	0.2514	Weifang City	0.2971	Linyi City	0.3793
4	Jinan City	0.2136	Weifang City	0.2341	Yantai City	0.2907	Weifang City	0.3788
5	Heze City	0.1695	Zibo City	0.2013	Linyi City	0.2532	Yantai City	0.3543
6	Weihai City	0.1614	Linyi City	0.1869	Heze City	0.2102	Heze City	0.2899
7	Linyi City	0.1587	Weihai City	0.1768	Weihai City	0.2042	Jining City	0.2511
8	Dezhou City	0.1475	Liaocheng City	0.1743	Jining City	0.1976	Dezhou City	0.2501
9	Jining City	0.1357	Heze City	0.1606	Dezhou City	0.1959	Liaocheng City	0.2398
10	Rizhao City	0.1344	Jining City	0.1564	Rizhao City	0.1771	Weihai City	0.2387
11	Dongying City	0.1222	Rizhao City	0.1547	Liaocheng City	0.1757	Dongying City	0.2047
12	Liaocheng City	0.1216	Dezhou City	0.1546	Dongying City	0.1652	Binzhou City	0.1993
13	Binzhou City	0.1139	Dongying City	0.1361	Zibo City	0.1627	Tai’an City	0.1965
14	Zibo City	0.1050	Binzhou City	0.1216	Binzhou City	0.1525	Rizhao City	0.1949
15	Tai’an City	0.0933	Tai’an City	0.1102	Tai’an City	0.1461	Zibo City	0.1898
16	Zaozhuang City	0.0752	Zaozhuang City	0.0819	Zaozhuang City	0.1042	Zaozhuang City	0.1488

shows the comprehensive score and ranking of agricultural digitalization construction readiness evaluation of cities in Shandong.

The change trend of the development level of agricultural digitalization in cities of Shandong Province from 2014 to 2020 is shown in Figure 3. As can be seen from the chart, the specific characteristics are as follows:

(1)

The overall development level of agricultural digitalization in different cities is on the rise, but there are obvious differences in the average values of different cities. The level of agricultural digitalization in Qingdao, Yantai and Weifang is obviously higher than that in other cities. The level of agricultural digitalization in Binzhou, Tai ‘an and Zaozhuang is obviously lower than that in other cities. By region, the average value of agricultural digitalization in eastern Shandong was the highest, followed by the middle of Shandong, and the lowest in southern Shandong.

(2)

The development level of agricultural digitalization in higher-value areas was unstable. The top cities (Weifang, Yantai, Heze, etc.) show a “concave” trend with lower middle and higher sides. In addition, the development of Jinan, Qingdao, Linyi great changes, the level of agricultural digitalization promotion speed; The development level of Yantai is stable, with a high-value fluctuation between 0.23 and 0.33.

(3)

Lower value areas fall into the “low level trap”. Low-value regions refer to the cities whose average level of agricultural modernization development is lower than 0.2 and ranking in the bottom three. The last three cities (Binzhou, Tai ‘an and Zaozhuang) have not been significantly improved, and the average level of agricultural modernization development of the above cities is lower than 0.2, unable to break through the state of stagnation at a low level.

4.4. Analysis of Obstacles in Different Stages of Agricultural Digitalization

According to Formulas (8)–(10), the barrier values of each secondary index of agricultural digitalization level in Shandong Province during 2010–2020 were calculated, and the results are shown in

Table 3. Obstacle values of various indicators of agricultural digitalization from 2014 to 2020.

Index	2014	2015	2016	2017	2018	2019	2020
C1 Rural electricity consumption	4.02%	4.78%	0%	5.11%	9.84%	11.16%	14.19%
C2 Average salary of Employees in Information transmission, Software and Information Technology Services	2.86%	3.14%	3.04%	2.49%	1.93%	2.09%	0%
C3 Local financial Expenditure on Science and Technology	9.82%	11.18%	11.82%	9.18%	7.71%	0%	7.2%
C4 Per capita expenditure on general public services	9.63%	0%	11.58%	10.11%	10.05%	9.06%	8.56%
C5 Total power of agricultural machinery	0.42%	0%	8.14%	7.36%	8.35%	9.48%	11.54%
C6 Effective irrigated area	5.59%	5.53%	5.15%	4.26%	3.6%	1.09%	0%
C7 Agricultural plastic film usage per unit area	7.77%	9.46%	0%	11.54%	15.15%	20.11%	27.56%
C8 Agricultural diesel oil quantity	0%	0.55%	1.44%	2.65%	6.55%	12.05%	22.7%
C9 Number of mobile phone users	6.01%	4.97%	4.92%	3.75%	2%	0.67%	0%
C10 Internet Broadband access Users	5.69%	6.52%	4.4%	3.56%	3%	1.65%	0%
C11 Expressway mileage per person	4.04%	4.41%	3.93%	3.85%	4.21%	3.89%	0%
C12 Computers per 100 population	4.53%	4.96%	5.07%	4.66%	3.69%	2.43%	0%
C13 Total Postal Services	10.14%	11.41%	11.17%	9.77%	9.45%	7.14%	0%
C14 Total telecommunications business	11.02%	12.92%	15.33%	13.9%	10.82%	5.64%	0%
C15 Per capita Gross output value of agriculture, forestry, animal husbandry and fishery	4.34%	4.57%	3.43%	4.11%	3.64%	3.54%	0%
C16 E-commerce Turnover of agricultural products	14.11%	15.58%	10.58%	3.69%	0%	9.55%	8.25%

. The larger the percentage of the barrier value, the higher the degree of constraint on the index. A percentage of 0 means that the constraint degree of the selected index is the least, but it does not mean that the index does not affect the development of agricultural digitalization.

As can be seen from Table 3, obstacles in agricultural digitalization in Shandong Province gradually show a two-level differentiation state, that is, obstacles of high value are more and more concentrated, and obstacles of other indicators are closer to 0.

(1)

In the deceleration and rise stage (2014–2015), e-commerce transaction volume of agricultural products (C16), total telecommunications business volume (C14), and total postal business volume (C13) are important factors restricting the development of agricultural digitalization. At this stage, Shandong Province will strengthen the interconnection between the existing agricultural and rural comprehensive information network resources, accelerate the construction and popularization of rural broadband network, and carry out the activity of “optical fiber into villages”. A number of “digital agriculture” demonstration projects have also been focused on, and the digitalization of agriculture is in its initial stage and making steady progress.

(2)

In the stable fluctuation stage (2016–2017), the total amount of telecom businesses (C14) and postal businesses (C13) are still important factors restricting the development of agricultural digitalization in Shandong Province. At this stage, the development of agricultural digitalization was stable, and the per capita general public service expenditure (C4) also had a major impact on the construction of agricultural digitalization. In 2016, the state vigorously promoted the development of big data in agriculture and rural areas, and promoted the application of Internet of Things technology in planting, animal husbandry, and fishery production to form big data of agricultural Internet of Things.

(3)

In the high-level rise stage (2018–2020), the amount of agricultural plastic film per unit area (C7) has become the most important restriction factor. Shandong Province attaches importance to ecological agriculture and advocates the organic combination of ecological protection and high-quality agricultural development. Soil digitalization is an important link in the development of digital agriculture and an important practice of the concept of “two mountains”. At this stage, rural electricity consumption (C1) and agricultural diesel oil (C8) both reflect the important influence of ecological digitalization on the development of agricultural digitalization. At the same time, in the Digital Agriculture Rural Development Plan (2019–2025) issued by the Ministry of Rural Agriculture in 2019, it was mentioned that the construction of basic data resource system should be focused on, the construction of digital production capacity should be strengthened, and the modernization of agriculture and rural areas should be driven by digitalization to provide strong support for the realization of comprehensive rural revitalization.

4.5. Analysis of Spatial Distribution Characteristics of Agricultural Digitalization Level

This paper explores the spatial distribution characteristics of agricultural digitalization development level in Shandong Province by comparing the data from different years. Using the ArcGIS10.8 software, the natural breakpoint method was used to generate the spatial change map of the agricultural digitalization development level in Shandong Province, and the agricultural modernization development level was divided into four levels: higher level, high level, low level, and lower level (Figure 4). According to the classification of agricultural digitalization development level, the spatial distribution of agricultural digitalization development level in Shandong Province has the following characteristics:

(1)

The proportion of the number of high-level cities increased from 6.25% in 2014 to 12.5% in 2020, and there was a certain spatial coupling with the regions with higher economic development levels. Qingdao was a high-level region at three time points. Qingdao seized the “new wind port” of digital agriculture and gave play to the role of seeds and data as basic and strategic resources. Leading the high-quality development of agriculture and rural work, the digitalization level of agriculture has been ranked first in Shandong. In 2020, Jinan will be transformed from a high-level provincial to a higher-level region. Relying on its resource endowment and agricultural characteristic industrial foundation, Jinan will promote the deep integration of cloud computing, Internet of Things, and artificial intelligence technologies with agriculture, which will continuously develop the digitalization level of agriculture.

(2)

The number of cities with high levels did not change, and most of them were concentrated in the eastern coastal areas. Weifang and Yantai always belonged to the regions with higher levels at three time points. In 2020, Linyi was upgraded from a lower level to a higher level, making the number of higher-level prefectures unchanged. Yantai is innovation-driven, gradually improves the modern agricultural system, and promotes the development of a digital economy enabling industry; Weifang has made important progress in the construction of digital agriculture and rural areas. The big data platform of “agriculture, rural areas, and farmers” has been completed, and the integration of big data, cloud computing, the Internet of Things, and other information technologies with all links of agricultural production and operation, management, and service as well as all fields of rural economy and society has been accelerated. In recent years, Linyi has continuously promoted the construction of agricultural information and promoted the transformation and upgrading of the agricultural industry.

(3)

Low and lower level areas are mainly concentrated in the northwest and south of Shandong Province around Jinan, and the number of prefectures and cities is also declining. The proportion of agricultural output value in most of these areas is high, but the level of agricultural digitalization is not high, and the degree of agricultural digitalization is far lower than the average level. Rizhao and Dongying also show the phenomenon of “back to low”.

4.6. Diagnosis of Agricultural Digitalization Level Obstacles

Obstacles to the digitized development of urban agriculture in Shandong Province in 2014, 2017, and 2020 were calculated, and the top two key obstacles in each city were obtained and ranked (

Table 4. Key obstacles of single index of agricultural digitalization construction in Shandong Province.

City/Year		2014		2017		2020
Higher value area	Qingdao City	C16	C14	C14	C7	C7	C4
		16%	12%	14%	14%	18%	15%
	Jinan City	C16	C14	C16	C14	C16	C7
		13%	11%	13%	10%	18%	14%
High value area	Linyi City	C14	C16	C16	C14	C16	C3
		14%	10%	14%	10%	18%	12%
	Weifang City	C16	C14	C16	C4	C4	C8
		15%	12%	15%	11%	13%	12%
	Yantai City	C13	C14	C16	C7	C7	C2
		15%	11%	14%	10%	12%	12%
Low value area	Heze City	C14	C12	C16	C14	C16	C3
		13%	11%	13%	11%	15%	11%
	Jining City	C10	C16	C16	C4	C14	C7
		13%	11%	13%	7%	12%	10%
	Dezhou City	C16	C14	C16	C14	C16	C4
		14%	11%	13%	11%	15%	10%
	Liaocheng City	C14	C13	C14	C3	C14	C3
		13%	11%	11%	9%	12%	11%
	Weihai City	C16	C10	C14	C7	C16	C7
		14%	11%	14%	9%	15%	10%
Lower value area	Dongying City	C16	C11	C16	C11	C14	C13
		13%	8%	13%	8%	13%	10%
	Binzhou City	C14	C16	C16	C14	C16	C7
		13%	10%	13%	10%	14%	9%
	Taian City	C16	C14	C16	C14	C14	C4
		12%	10%	13%	10%	10%	9%
	Rizhao City	C13	C16	C16	C7	C16	C4
		13%	10%	13%	9%	14%	7%
	Zibo City	C16	C1	C16	C14	C16	C7
		13%	10%	13%	11%	13%	9%
	Zaozhuang City	C16	C14	C16	C4	C16	C3
		12%	10%	12%	8%	13%	9%

).

As a whole, the major obstacles affecting the cities are the e-commerce transaction volume of agricultural products (C16) and the total volume of telecommunications business (C14). The use of agricultural plastic film per unit area (C7) in the higher level areas was also significantly affected in the later period, indicating that the development of ecology played a certain role in the construction of agricultural digitalization. The per capita expenditure on general public services (C4), the average salary of employees in information transmission, software and information technology services (C2), and the amount of agricultural diesel (C8) have a great impact on some high-value cities. In low-level and low-level regions, local financial expenditure on science and technology (C3) plays an important role in many low-value regions, indicating that the national financial support plays a significant role in this region.

5. Conclusions and Policy Implications

5.1. Conclusions

In this paper, by constructing the evaluation index system of agricultural digitalization development, the entropy method is adopted to comprehensively evaluate the development level of agricultural digitalization in Shandong Province, analyze its development trend and internal structural changes, natural breakpoint method is adopted to analyze regional differences of agricultural digitalization level in Shandong Province, and obstacle degree model is used to diagnose the main factors restricting the development of agricultural digitalization. Through the above results, the main conclusions are as follows:

(1)

From the time dimension, the agricultural digitalization readiness of Shandong Province can be divided into three stages: the deceleration and rise stage (2014–2016), the rapid rise stage (2016–2018), and the high-level fluctuation stage (2018–2020). The high value of the agricultural digitalization development level index fluctuates, while the low-value area falls into the “low-level trap”. In order of time, most of the cities in the middle and upper reaches of Shandong Province fluctuated with time. However, the average ranking of the cities in the lower reaches of the development level of agricultural digitalization has been stagnant at a low level, and the improvement is not obvious.

(2)

At the spatial level, the agricultural modernization level of Shandong Province presents the spatial differentiation characteristics of high in the east and low in the west, which is consistent with the conclusion that the development level of digital agriculture in China presents the spatial distribution pattern of “east-middle-west” from high to low. The high-value areas are distributed in the eastern coastal areas, and their secondary indexes are mostly at the forefront of the province. The regions with high values are mostly distributed in the surrounding areas centered on Qingdao, while the regions with low values are mostly distributed in the western and southern parts of Shandong, and the regions with low values are mostly locked in the surrounding areas of Jinan.

(3)

At the level of obstacle degree, high-value obstacles are gradually concentrated. In different stages of agricultural digitalization development, the main obstacles are different. In each stage, there is a most important factor playing a role. For each city, the e-commerce transaction volume of agricultural products and the total amount of telecommunication business become the main obstacle factors. In addition, the obstacle factors of cities at the same level are similar, but there are differences in the obstacle factors of cities at different levels.

5.2. Policy Implications

The study presents empirical results and key conclusions and proposes countermeasures and suggestions to improve the development of agricultural digitalization across different regions and achieve comprehensive and coordinated growth. The following measures are suggested:

• Accelerate the development of a comprehensive digital agriculture system that encompasses the entire industrial chain of agriculture. This will involve connecting the upper, middle, and lower reaches of the agricultural industry chain and integrating digital technology and services throughout the agricultural and rural work processes. Local governments should adapt to local conditions to promote high-quality development of agriculture and rural areas.
• High-value regions should explore long-term mechanisms to drive digitalization and support high-quality agricultural development. This can be achieved by building and improving government policy support, market drive, collaborative research and development, and participation of intermediary service institutions. Digital technologies such as the Internet of Things, remote sensing observation, and navigation and positioning should be integrated into the agricultural industry to improve the quality and efficiency of agricultural information.
• Low-value areas should improve their digital environment by accelerating the promotion of “new infrastructure” in rural areas, such as broadband access, rural e-commerce, and logistics. Innovative models should be used to promote the development of non-physical products and services in rural areas.
• Strengthen inter-regional exchanges and cooperation to cultivate the growth poles of low-value areas. Regional exchanges in digital technology and management experience should be encouraged, and the extension of digital applications should be explored. Low-value areas can also promote the return of agricultural digital technology talents by increasing the financial transfer payment.

The study concludes that the advancement of science and technology is crucial for modernizing agricultural production facilities and achieving intelligent management of the entire agricultural production process. However, measuring agricultural digitalization remains a challenge, and future research will focus on addressing this issue and accelerating the digital transformation of agriculture.