**1. Introduction**

Consumers are no longer satisfied with having enough to eat; food quality has become a key factor in determining consumer choice [1], and their demands and preferences change with the season, time, weather, mood, and other factors [2]. However, food choice is a luxury not every person enjoys. The Food and Agriculture Organization (FAO) of the United Nations reported that 88% of countries face a serious malnutrition burden and so has issued healthy dietary guidelines that cover a wide range of food and nutrition (http://www.fao.org/nutrition/education/food-dietary-guidelines/ regions/countries/united-states-of-america/en/(accessed on 2 September 2021)). Traditional food science has been unable to satisfy increasing demand for food in a world where "healthy nutrition" has overtaken "well-fed" as the predominant paradigm of consumption (https://baijiahao.baidu.com/s?id=1681293145359468742&wfr=spider&for=pc (accessed on 2 September 2021)) [3]. However, big data offers food science a new means of scientific analysis [4].

The food supply chain is composed of economic stakeholders from primary producers to consumers. It has the characteristics of large volume, many links, wide distribution, diverse types, and scattered data, and it is becoming more complex. Millions of tons of food move around the world every year, so no enterprise can promise that every risk node on the production line is absolutely safe. Any flaw in the supply chain could bring a disaster and huge regulatory difficulties for governmen<sup>t</sup> departments. However, big data provides a solution to regulatory difficulties [5] by helping enterprises understand consumer demand better and uncover food industry trends through big data analysis. The food industry collects large datasets through real-time monitoring and can improve food safety if analyzed in conjunction with sample data [6]. When industry data is combined with data on consumer dietary behavior, food enterprises can optimize their investment and adjust the direction of research and development in a timely manner. [7].

This paper uses bibliometrics to analyze the research progress of big data in the food field. According to Bradford's Law, a small number of core journals collect enough information to reflect the latest and most important advances in science and technology. The database of Web of Science Core Collection contains more than 12,000 core journals

**Citation:** Tao, Q.; Ding, H.; Wang, H.; Cui, X. Application Research: Big Data in Food Industry. *Foods* **2021**, *10*, 2203. https://doi.org/10.3390/ foods10092203

Academic Editor: Maria Lisa Clodoveo

Received: 19 August 2021 Accepted: 11 September 2021 Published: 17 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

from more than 250 subject areas. It defined the search topic "Food & Big Data" and selected 1672 papers from its database. The research progress of big data on food is shown in Figure 1 It has increased significantly since 2014 because USD 35.8 billion was invested in global agrifood from 2010 to 2019, and after 2014 the scale of financing grown rapidly (https://agfunder.com/research/agfunder-agrifood-tech-investing-report-2019/ (accessed on 2 September 2021)). This increased capital investment promoted research into big food data, and the rapidly rising trend is from 2010 to 2021 is shown in Figure 2. China's food industry has attracted global attention since 2012 because the country's new governmen<sup>t</sup> leaders stressed that they would pay more attention to food safety (http: //www.xinhuanet.com/politics/2017-01/03/c\_1120239001.htm (accessed on 2 September 2021)). As the second largest economy and the world's largest trading country, China has grea<sup>t</sup> international influence. (https://www.brookings.edu/research/chinas-influenceon-the-global-middle-class/ (accessed on 2 September 2021)). Big data has been one of the focuses of research since 2013, mainly in food safety, food security and agriculture. Its application to food safety may still be in its infancy, but it is affecting the entire supply chain. The literature contains analyses on the feasibility and need for big data in the food industry [4,6], but there are no in-depth analyses. Therefore, this paper will mainly discuss the following three aspects in depth.


•

**Figure 1.** Research progress of data in the food field. From 1990 to 2010, the research papers of big data on food grew at a rate of 100% every five years, and since 2010 it has grown by nearly 300% every five years.

**Figure 2.** Analysis of the research direction of the data from 2010 to 2021 in the food field. Since 2013, food security and big data have become a focus of researchers interested in the potential value of big data on food. The research focuses on IoT-based data collection and its application to smart farming, supply chain management, food nutrition, and sustainable development.

The authors of this paper hope to help researchers develop a deeper understanding of the research progress of big data in the food field and to provide guidance for further research.

### **2. Big Data in Food Industry**

Big data sources of food mainly include regulatory, food enterprise (including data generated at every link of the industrial chain from planting to restaurants) , and media data (including food-related news, video, pictures and audio). High-quality big data analysis can help develop the food industry, whereas analyses from low-quality data can adversely affect managers' prediction of market demand [8], and social stability [9].

### *2.1. Food Regulatory Data*

Food regulatory data usually includes department regulatory and product sampling data. Marvin [4] has detailed public information about food safety supervision and sampling inspection in various countries. This information includes, reports on animal and plant disease monitoring, hazards, food-borne diseases, which provided support for researchers of deep-risk information. Rapid Alert System of Food and Feed (RASFF) is a commonly used online food safety database for industry and scientific research in the European Union (EU). Food safety databases in other countries include the Import Rejection Report (IRR) and the Inspection Classification Database (ICD) in the U.S. and the State Administration for Market Regulation (SAMR) alerts in China [5]. With the increasingly close connection between countries, the trend of "table globalization" has become increasingly prominent. In 2017, the amount of food China imported from Australia, the United States, Japan, Germany, Southeast Asia, and other countries exceeded RMB 1.5 trillion (https://www.askci.com/news/chanye/20171212/084457113784.shtml (accessed on 2 May 2021)). As shown in Table 1, the SAMR usually shares its sampling inspection results of imported and exported food on governmen<sup>t</sup> websites, which allows consumers to know the quality of food on the market. The U.S. governmen<sup>t</sup> shares food sample analysis reports through the FSIS system. The EFSA database contains data on food consumption habits and patterns across the European Union. Such statistical data allows users to quickly screen long-term and acute exposures to potentially hazardous substances in the food chain. In addition, the World Health Organization (WHO) established the Global Environmental Monitoring System (GEMS/Food) in 1976, in which participating institutions submit data on food pollutant concentrations and set up data centers to help governments, the Codex Alimentarius Commission(CAC) and other institutions to assess trends

in food contaminants [10]. In 2015, the WHO integrated data from the fields of agriculture, food, public health and economics to build a big data services platform for food safety (https://www.who.int/foodsafety/foscollab/en/ (accessed on 2 May 2021)) to improve risk monitoring.


**Table 1.** The public regulatory database.

**Challenges.** The sharing and circulation of data among food regulatory departments is conducive to the construction of intelligent supervision of the food supply chain [11].However, there are several challenges.


### *2.2. Food Enterprise Data*

The food industry chain is composed of enterprises from agriculture, fishing, processing and restaurant and is characterized by many links and wide distribution. At present, all agricultural machinery is electronically controlled to improve operational performance [14,15]. Cloud computing, the Internet of Things, big data and blockchain can integrate isolated production lines in the food supply chain into data-driven interconnected intelligent systems. Through semantic active technology, each operation is automatically integrated, improving the efficiency of precision agriculture and enterprise managemen<sup>t</sup> [16]. Using sensors and drones to collect data on weather, geography, and animal and crop behavior can help farmers optimize crop planting and animal growth cycles. Intelligent devices capture actionable data and make decisions that reduce equipment downtime [17].

In recent years, research on the IoT in the food industry has promoted the diversification of the IoT platform to address market needs, [18,19] different monitoring models [20], and unbalanced energy consumption [21]. IoT-integrated applications will help food companies create new data sources. Industry 4.0 not only promotes the rapid development of Agriculture 4.0, but also enables enterprises to transmit real-time information to identify and meet the changing demands of stakeholders [15]. According to the Eurostat report (https://www.brookings.edu/research/chinas-influence-on-the-global-middle-class/ (accessed on 2 September 2021)), the application of smart agriculture will save 4–6% of agricultural costs and increase market value by 3% by 2026. The application of big data can not only enable businesses to deal with challenges in food production, but also to obtain more affordable raw materials to reduce production costs [14]. It also promotes the development of smart agriculture, which helps save water [22], preserve soil, limit carbon

emissions [23] and improveproductivity [24]. Smart agriculture provides an opportunity for farmers, service providers, governmen<sup>t</sup> and other stakeholders (such as financial institutions, investors, traders) to share their experiences in optimizing the agricultural supply chain with the production sustainability [25].

**Challenges.** The food industry can benefit from big data services, but there are challenges that need to be addressed, including data fairness such as the searchability, accessibility, interoperability, and reusability of shared data, and a lack of information standards and data processing technology.

