*Article* **The Role of Logistics in Food Waste Reduction in Wholesalers and Small Retailers of Fruits and Vegetables: A Multiple Case Study**

**Patrícia Guarnieri 1,2, Raiane C. C. de Aguiar 2, Karim M. Thomé <sup>2</sup> and Eluiza Alberto de Morais Watanabe 1,\***


**Abstract:** *Background:* There is a lack of studies on the waste of fruits and vegetables covering both distributors and the retail sector. Our study advances from previous ones by proposing the analysis of local retailers of different sizes and wholesalers. Our objective was to analyze the logistical practices to reduce the waste of fruits and vegetables in wholesalers and small retailers in Federal District, Brazil. *Methods:* A multiple case study was conducted with 19 retailers and eight wholesalers. We administered semi-structured interviews and performed documental analysis and direct observation. *Results:* The findings demonstrated the leading logistical practices of handling, conservation, management and control, and consumer awareness about food waste. The type of retailer was limited to grocery stores and fruit shops, and the type of food covered only fruits and vegetables. Furthermore, we considered the food waste generated in the logistics processes and not that after consumption. *Conclusions:* More than a third of the food produced worldwide is lost or wasted. A considerable part of the food waste is associated with the lack of an adequate structure of the logistical processes. The results may contribute to the adoption of practices related to reducing food waste by retailers and wholesalers.

**Keywords:** food waste; logistics; retail; wholesale; supply chain management

#### **1. Introduction**

Despite the expeditiousness of world food production, the United Nations Food and Agriculture Organization [1] points out the need for a 60% increase in global food production by 2050 to meet the population's growing demands. Brazilian agricultural production has grown markedly in the last decade, facilitating the reduction of poverty and hunger in the country [2]. Specifically, in relation to fruit production, Brazil stands out as the third largest world producer, responsible for approximately 45 million tons of produce every year, with about 65% for domestic supply and the remaining 35% destined for exportation [3]. Additionally, the Brazilian vegetable chain presents various options, concentrating its production volume on the following species: potato, tomato, watermelon, lettuce, onion, and carrot [3].

Although the volume of food production in Brazil is enormous, a substantial part is wasted [4]. Brazil wastes around 41 thousand tons of food per year [5]. Worldwide, more than a third of the produced food is lost or wasted, equivalent to about 1.3 billion tons of food [6]. Among the amounts lost or wasted, 30% corresponds to cereals; between 40% and 50% comprises roots, fruits, vegetables, and oilseeds; 20% involves meat and dairy products; and 35% is fish. It is estimated that these foods would be enough to feed two million people [6]. Within this context, reducing food losses and waste should be a priority

**Citation:** Guarnieri, P.; de Aguiar, R.C.C.; Thomé, K.M.; Watanabe, E.A.d.M. The Role of Logistics in Food Waste Reduction in Wholesalers and Small Retailers of Fruits and Vegetables: A Multiple Case Study. *Logistics* **2021**, *5*, 77. https://doi.org/ 10.3390/logistics5040077

Academic Editor: Robert Handfield

Received: 13 July 2021 Accepted: 7 October 2021 Published: 4 November 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/).

for establishing productive structures and structures of sustainable consumption [1]. The irrational use of food harms humanity [7]. Understanding food waste is crucial as it affects food chains. For example, the European Union and other high-income countries have significantly higher levels of food waste [8]. In developing countries, such as Brazil, waste is mainly linked to the initial stages of the supply chain involving harvesting, transportation, storage, and distribution [9]. About 30% of the total vegetables that pass through the distribution centers are lost, with only 70% being marketed [10].

Analyzing the overview of worldwide studies, it is possible to verify an emphasis on the valuation of the final consumer behavior [11–13], the measurement of the wasted quantities [14,15], and the environmental and monetary impacts generated by food waste [16,17]. In Brazil, some studies have addressed food waste under the perspective of logistical bottlenecks [18]. The authors approached specifically the handling and transportation activities, verifying the rates of losses of vegetables and fruits. Other studies addressed the role of packaging and the supply chain's coordination structures to reduce food waste [19]. The use of food waste in a different production process was studied by Belik et al. [20] and Fagundes et al. [21]. Finally, some studies stressed the causes of food loss and ways to reduce food waste [22–24].

Considering this scenario, there is a lack of studies on the waste of fruits and vegetables covering both distributors and the retail sector. Thus, the present study advances from previous ones by proposing the analysis of local retailers of different sizes and wholesalers. This paper aims to analyze the best logistical practices in reducing the waste of vegetables and fruits in retailers and wholesalers located in the Federal District, Brazil. For this purpose, this study was composed of two parts: first we identified the causes of waste through a systematic literature review, considering Brazilian and international studies, and second, we carried out an analysis of multiple cases through direct qualitative interviews conducted with 19 managers/owners of fruits and vegetable stores and eight wholesalers, who supply the fruits and vegetables to the stores. The data analysis was carried out by means of categorical content analysis.

The main contribution of this paper is twofold: (i) the systematization of the best logistical practices of fruits and vegetables to avoid food waste employing a systematic literature review; (ii) the analysis of the best logistical practices to reduce the waste of fruits and vegetables adopted by retailers and wholesalers in the agri-food chain, in terms of handling, conservation, management and control, and awareness.

#### **2. Literature Review**

#### *2.1. Agri-Food Supply Chain and Food Waste*

The agri-food supply chain refers to a series of relationships in different segments that establish successive exchanges in transforming inputs into value for the final consumer [25].

The agri-food supply chain starts at the point "Before the Farms", covering the activities related to inbound logistics and companies that supply raw materials to be used in the production process in the field, such as seeds and agrochemicals. Then, there is the link "On the Farms", which considers the improvements that products still receive inside the farms: weighing, pre-washing, selection, cooling, agro industrialization, and packaging. The last link, which is the focus of this research, is the "After the Farms" phase, covering outbound logistics activities, including activities of handling, storage, warehousing, and transport. It considers attempts to sell products to industries and other distribution channels until they reach the final consumer [10].

Within the agri-food supply chain, specifically for fruits and vegetables, food waste can occur, mainly due to the high perishability of the product and incorrect handling and transportation. It is essential to point out that there are two main types of food waste. When there are losses until the process of distribution, we call them "food losses". After the distribution process in several channels (wholesalers, retailers, small stores, fairs, direct sales to consumers) and after the sales, at the consumers' residence, we call them "food waste". This study is focused on the food waste occurring in retailers and wholesalers. In

this process, many products that individuals can still consume are discarded. The waste of fruits and vegetables is mainly associated with the behavior of wholesalers, retailers, other food sales services, and consumers [6]. Food waste encompasses all food discarded in the marketing stages by retailers and wholesalers, in which the food is discarded under full consumption conditions since it would still meet nutritional needs [9].

It is noteworthy that a considerable part of the waste is associated with the lack of adequate structure of the logistical processes involved [4], which is the focus of this study. The way the supply chain components are structured directly influences the quality of the products [25]. Furthermore, errors in demand forecast, inefficient replenishment policies, and high product quality demand can contribute to food waste [26]. Structural problems, such as planning and logistical bottlenecks, limit improvements in Brazil's fruit and vegetable supply chain stages [27]. After the harvest, food waste in Brazil is caused by improper packaging, lack of product refrigeration, improper handling, poor display of products on the shelves, deficiencies in transportation, and incorrect handling by consumers [28].

In this sense, reducing food waste is a challenge that requires enhancing the efficiency of the operations and logistics [29]. It may include infrastructure and hygiene care advances and better management and conservation of the fruits and vegetables in the market [30]. Furthermore, improvement in logistics, cold chain management, retail packaging, and consumer awareness publicity can prevent food waste [31].

#### *2.2. The Best Logistical Practices to Reduce Waste of Fruits and Vegetables: Systematic Literature Review on Brazilian and International Publications*

In order to get the leading logistical practices aimed at reducing the waste of fruits and vegetables, we conducted two systematic literature reviews. This section presents the results of the literature reviews, on the basis of which category construction was carried out for data collection and analysis. The description of the used protocol to select and filter papers is included in Section 3, related to methodological procedures. Table 1 shows Brazilian and international publications' analysis of best practices in logistics to reduce/avoid food waste.


**Table 1.** Best practices in the logistics stages to reduce and/or avoid food waste.

**Table 1.** *Cont.*




According to Table 1, among the best practices can be highlighted the concerns to improve the quality of food, the extra care to handle fruits and vegetables, the proper packaging and facilities, such as structures with refrigerated chambers and shelves, management of the quantity offered, the information sharing and collaborative partnerships, and the education and awareness of the final consumer.

The main logistical actions to reduce food waste due to deterioration are purchase planning, storing products in air-conditioned environments, appropriate boxes and packaging, correctly exposing products, conducting campaigns with customers to properly handle products, acclimatizing the store, and regulating the store refrigerators and freezers [22]. The authors also proposed actions to avoid losses due to packaging damage: care in unloading and handling products, improving storage, training employees, and reinforcing care in the proper display of products. Some internal and external measures to reduce vegetable losses were pointed out by Tofanelli et al. [23]. The internal ones: improving inventory control, purchase of fresh vegetables, purchase of regional vegetables, decrease in retail prices, care in handling during transportation, prevention of excessive handling by the consumer, and improving the structure of the establishment. The external ones: lower wholesale prices, closer wholesale suppliers, educating the final consumer, enhancing the quality of vegetables, improving packaging, encouraging local vegetable growing, and greater integration and collaboration of members of the agricultural supply chain.

The measures to reduce food waste are cleaning transport, monoblock boxes, and marketing benches; uniformity in the organization of vegetables in boxes; selecting the best times for the outlets; and product offers according to demand [24].

The great majority of studies aim to understand the behavior of the final consumer, listing the factors that drive food waste, considering that some studies point out that, at the consumption stage, there is more generation of food waste and a greater possibility of prevention [11]. However, although consumers appear to make the most significant contribution to the wasted food volume, there is almost no information on the drivers of such behavior in consumer households [40]. In this sense, the minimization of food residues in developed countries should be focused on the retail and consumption stages [39]. The waste of consumer-related food is a complex issue requiring collaboration among various actors in the supply chain and actions to increase awareness [43].

#### **3. Research Techniques**

This research is characterized as applied, descriptive, and exploratory and uses a qualitative approach. We conducted a systematic literature review and a multiple case study. Table 2 shows the technical procedures adopted and their respective research instruments to collect data. The study was divided into two stages and related technical procedures. The first phase covered two systematic literature reviews, and the second one the multiple case study.


**Table 2.** Relation between technical procedures and data collection instruments.

We carried out two systematic literature reviews to demonstrate state of the art on food losses and waste. In addition to summarizing the main problems throughout the agrifood supply chain, the review pointed out solutions through the implementation of best practices. Two different protocols were used to carry out the systematic literature reviews: (i) the Brazilian one used the protocol from Cronin et al. [45] and (ii) the international one used the protocol from Pagani et al. [46]. Two procedures were necessary considering that the Brazilian journals, at most, do not have impact factor information, which would be required to calculate the InOrdinatio Index proposed by Pagani et al. [46]. Thus, the protocol from Cronin et al. [45] was used. To summarize, the first three steps were the same in both protocols; they differed in terms of the filtering process, because the Methodi Ordination uses the InOrdination as an additional step of filtering.

The Brazilian literature analysis followed the protocol of Cronin et al. [45], with the following steps: (a) research question formulation; (b) set of inclusion and exclusion criteria; (c) selection and access to literature; (d) quality evaluation of the literature included in the review; and (e) result analysis, synthesis, and dissemination. From the keywords food waste OR food losses, in the Portuguese language, 15,100 results were found on the Google academics platform in order of relevance. The titles and keywords of the first 400 links were selected and evaluated. Due to the filtering process, 78 were eliminated. In addition, 310 were excluded due to inconsistency with the scope of the study. After reading the abstracts, 12 articles were thoroughly analyzed.

In the systematic international review, the Methodi InOrdinatio was used to classify the quality of the articles [46]. The method allowed us to select the best articles by ordering the highest scores. The calculation considers the publication year, the impact factor of the journal, and the number of citations. Pagani et al. [46] advise the researcher to determine the cutoff line so that only the articles above the line are read in full. Thus, 17 articles were selected, which presented an InOrdinatio superior to 123. At first, 437 articles were found, considering the same keywords in the English language. The studies were analyzed by reading the titles and abstracts. Of the 437 articles, 358 were excluded due to inconsistency with the theme. The remaining 79 articles passed through the Methodi Ordinatio filter, which calculates an index called InOrdinatio to rank the papers considering the number of citations, year of publication, and impact factor, leaving, in the end, 17 articles higher than the cutoff line of 123 points of InOrdinatio. Thus, the final systematic review included 29 articles—12 Brazilian and 17 international papers. Figure 1 presents the results of the two SLRs. It is important to emphasize that this procedure supported the questionnaire elaboration applied with the sample of participants of the study.

**Figure 1.** Protocol to select and filter papers analyzed in the SLR.

For this reason, the SLR considered just papers published until 2017. We understand that this is a limitation of this study. Further studies can confirm the logistic practices from 2018 to the future.

Regarding the second phase of the research, the study of multiple cases was carried out mainly through semi-structured interviews and by the direct observation "in loco" in the stores of fruits and vegetables located in the Federal District and wholesaler storages, with the professionals responsible for the fruit and vegetable commercialization (owners and managers), involving retailers and wholesalers in the Federal District, Brazil. The study was carried out at Ceasa/DF, the center of food distribution (wholesale) and local retailers, located in the following administrative regions of the Federal District: Samambaia, Taguatinga, Ceilândia, Guará, and Asa Norte. The interviews were conducted in person, at the establishments, without an appointment by occasional visits, with an average duration of 25 min. The interviews were carried out from 2019 to 2020 and recorded with the authorization of the interviewees to allow the transcription for further interpretation and analysis.

Nineteen fruit and vegetable retailers and eight wholesalers from the CEASA-DF were part of this analysis. It is noted that a sample was not established a priori, being followed by the theoretical saturation technique. Regarding the size of wholesale establishments, 37.5% were large companies, and the remainder were small ones. Of the retailers surveyed, 16% were individual microentrepreneurs, 53% were microenterprises, 21% were small, and 10% represented large companies. Regarding the profile of the interviewees, 53% were the owners of the establishments, 26% were managers, 16% were replenishment managers and, 5% were sales managers. Table 3 shows the characteristics of the study participants.


**Table 3.** Characterization of participants of the study.

According to Table 3, 19 companies were part of this analysis, representing the retail sector of fruits and vegetables, 26% located in Samambaia, 21% in Taguatinga, 11% in Ceilândia, 27% in Guará, and 15% in Asa Norte. The companies were found in searches conducted on the Google Maps tool, based on the search for keywords in Portuguese: "Verdurão," "sacolão," "frutaria," and "Hortifruti," which are terms used to refer to retailers specialized in fruits and vegetables in Brazil. Thirty-six establishments were found, of which six were no longer in operation, four did not accept to participate in the study, and the other seven were not accessible—four due to their location and the remaining three because they belong to foreigners who are not able to communicate in Portuguese preventing the conduction of the interviews. Therefore, the choice of the participating companies followed the criteria of accessibility (the managers/owners should agree to participate) and representativeness (the companies should meet the search criteria).

A semi-structured interview script, designed on the basis of the systematic literature reviews, was used for data collection. The script went through analysis by judges to give more robustness to the items. The judges were seven teachers with an affinity to the themes of logistics and agribusiness. The questions were analyzed for the criteria of intelligibility, clarity of information, and coherence of terms, considering the pre-established objectives. The script was composed of four categories: Handling; Conservation and Maintenance; Control and Logistics Management, and Awareness. In total, 23 questions were added, with six items on the first category, seven items on the second category, nine items on the third category, and one item on category 4.

We conducted a thematic categorial content analysis technique to analyze the results, following the protocol proposed by Bardin [48]: pre-analysis; exploration of the material; and treatment of the results, inference, and interpretation. We detailed the results in categories that are analyzed under the thematic content of the interviews and documental analysis, enabling the identification of the meaning in the interview composition. The categories established a priori were as follows: (a) handling, (b) conservation and maintenance, (c) control and logistical management, (d) awareness.

#### **4. Results and Discussion**

Table 4 presents the main obtained results from interviews and direct observation for the categories handling, conservation, management and control, and awareness for the local retailers and wholesalers of fruits and vegetables considered in this study.





The handling category identified the receiving, storage, warehouse, and commercialization practices (Table 4). As soon as the trucks arrive at the retail establishment, manual unloading or unloading with the aid of trolleys and pallet trucks is made, which facilitates handling work and reduces injuries due to impact, upon receiving the cargo, checking what is being received stands out regarding the quantity (weight) and the products' quality (uniformity, degree of maturation, and presentation of injuries or imperfections). The storage and warehousing processes involve stacking the boxes, removing at first the excess of goods at the edges of the boxes.

In 40% of the establishments that have storage for the fruits and vegetables, there are no cold rooms to store them, which reduces their durability. All establishments reported having significant waste due to excessive and incorrect handling by the final consumer (e.g., kneading and squeezing food). However, 27% of establishments did not advise consumers on the correct handling of fruits and vegetables to avoid embarrassment.

The conservation category identified the practices of conservation, maintenance/hygiene, organization, and separation of food. The increase in durability and the preservation of the quality of the exposed items under refrigeration stood out, which considerably reduces waste. The use of packaging is essential for the conservation of fruits and vegetables, which suffer effects related to the transport itself and the sales environment, and the effects of excessive handling. In turn, sanitation impacts the durability of fruits and vegetables concerning the removal of insects.

The shelves' organization is relevant in terms of separation by type and group of products. Damaged fruits and vegetables are removed when the shelf is replenished and throughout the day to avoid the spread of degrading agents. Products considered imperfect, outside the aesthetic standards required by customers, are removed from the shelves or are not exposed. Then, they are directed to the employees' consumers, avoiding waste—some establishments process these imperfect products for sale as fruit pulps, soup preparations, fruit salad, and natural juices.

The management and control category identified practices related to demand forecasting and the use of inventory management tools, a strategy to increase product turnover, and measures to reduce the distance between retail and supplier, in addition to the reverse logistics of waste and imperfect food. The results showed the absence of purchase planning and predictions made through notes and lists without strategic planning to reduce waste. Excessive purchases generated waste of fruits and vegetables even for establishments that make daily purchases. It was noted that the adoption of information technology tools for the planning of purchases did not occur due to the cost, the spent time, and the owner's refusal.

The strategies to increase the product turnover revolved around promotions (for example, Green Tuesday and Green Wednesday) and the offer of a single price, the socalled *sacolão*. The practice of *sacolão* has become unfeasible due to price discrepancies and the variety of products offered. We observed that there was little exchange of information between retailers and wholesalers about the demand forecast. As a result, the wholesaler's planning was also affected. Negotiation with suppliers of items with minor imperfections was a possibility given waste mitigation strategies. The reduction of waste also occurred through the use of foods that were still within the expected nutritional level but were no longer considered within the standards of retail marketing because they contained some imperfection or are damaged in some parts.

Regarding the possibility of food donation, there was difficulty due to the cost of transporting the products to the entity to be benefited. Food unfit for human consumption was used for animal feed or compost production or was discarded in the garbage for collection by the urban cleaning service. Regarding the awareness category, none of the retail establishments had awareness campaigns for the final consumer. The impediments revolved around the costs and spent time. Moreover, retailers pointed out that they had never been approached by the private sector or government institutions for presentation and incentives for initiatives in this regard.

Regarding the study carried out with wholesalers, three categories were established. The first was about handling the receiving, storage, and marketing stages. The results show that about 50% of the food sold came from the northeast and southeast regions of the country, 25% from the south, and 12.5% from the central-west region. In companies that work with imported products, about 12.5% came from Argentina and Chile. Regarding the unloading process of the trucks, the loads were received, weighed, and taken into the box. During the receipt of the products, the employees responsible for checking the loads evaluated the temperature, uniformity, degree of maturity, and quality of the received food. Regarding batch uniformity, fruit sizes were checked, standardization was always prioritized, and the most presentable items were separated from imperfect ones. The cold chambers located inside the box were used so that the fruits were packaged and had prolonged durability and a visible reduction of waste.

The second category was the conservation, maintenance/hygiene, and separation of damaged or imperfect items. In terms of conservation in the commercial environment, we highlight the establishments that used cold rooms. It is noteworthy that CEASA stipulates that permit holders use correct packaging without using wooden boxes. However, the use of these is still being verified. The issue of transporting cargo over long distances was one of the factors to be analyzed since conservation requires refrigerated box trucks. However, the fleet varied widely between open bodies, closed box trucks, and refrigerated box trucks.

Regarding the maintenance and hygiene of the commercialization environments, the licensees cleaned the box and washed the containers frequently. Regarding the use of insecticides for the removal of insects, the licensees informed us that CEASA was responsible for environmental pest control. Concerning removing damaged or imperfect products, the interviewees aimed to remove damaged products and discard them so that they did not damage the cargo. For defective items, just as in retail, price offers were the first choice. Another mentioned option was sale to establishments that would process this food, such as snack bars and restaurants.

Furthermore, the third category referred to demand forecasting and the use of inventory management tools, a strategy to increase product turnover, the relationship between retail and supplier, and CEASA's role in terms of waste and reverse logistic waste and imperfect food. Wholesalers who used a management system that assisted in planning the purchase had difficulty interpreting the information. This highlights the importance of training and development of employees, not only regarding correct handling but also with respect to feeding information systems. It should be noted that the large stocks held were responsible for a large part of the waste at the supply terminals. To increase the turnover of products, the sale of products to the "sacolão" was pointed out, the negotiation of a lower price with the producer, the search for customers and the guarantee that they work with a quality product.

It was noticed that the wholesaler link, the permit holders of CEASA-DF, usually exchanged price information and the quality of the products with the producer. The information exchanged between retail and wholesale revolved around the requirements of retailers concerning quality, price, delivery time, and merchandise exchanges. There was no sharing of strategic information among members about forecasting demand or stock level to reduce waste. Respondents pointed to the zero-waste program as the initiative to reduce waste. In addition to sending damaged and imperfect food to CEASA-Federal District's zero-waste program, respondents reported making donations to entities such as daycare centers, churches, non-governmental organizations (NGOs), and SESC (Social Service of Commerce). There was no waste reuse program for composting or producing animal feed. There was still much waste at CEASA-DF due to the lack of awareness and involvement of permit holders in waste reduction practices. Some studies deal with solutions to improve food donations from restaurants to the food-insecure population using modeling [49], which can be pointed out as a positive effort involving the storage, collection, and transportation logistical activities to reduce food waste.

#### **5. Conclusions**

This study dealt with the analysis of good logistical practices in reducing fruit and vegetable waste in retail and wholesale companies, as these marketing links comprise stages in the supply chain with a high percentage of waste. To provide a broad and diversified panorama and to get inputs for the elaboration of the questionnaires to collect data, a systematic national and international review was carried out, which supported the item's construction of the semi-structured script used in the interviews in the study of multiple cases, which is the central part of this study.

Given the obtained results, the need to make consumers aware of the correct handling of fruits and vegetables in retail establishments is highlighted. In addition, we emphasize the requirement for the companies to carry out campaigns with public–private partnerships to educate the consumer, either in terms of purchase planning as in the full use of food. Regarding the wastes that still occur in CEASA-DF, measures to raise the awareness and involvement of permit holders must be adopted, as well as the development of facilities for processing the fruit and vegetables that end up being discarded in containers and destined for landfills.

The research has several contributions. Academically, considering the importance of studying the proposed theme in depth, the research contributes to a more significant discussion about food waste and the investigation of the phenomenon in the retail and wholesale environment. Specifically, the research contributed to the panorama of the national and international state of the art about the good practices carried out along the agricultural supply chain. Furthermore, the logistical practices adopted by two links in the agri-food chain, retailers and wholesalers, were identified to reduce food waste.

From a managerial point of view, the study contributes to retail establishments creating employees and consumers awareness campaigns as to the problems caused by excessive and inadequate handling in the receiving, storage, and marketing stages, contributing to the structuring of marketing processes. As well as in identifying the variables necessary for the maintenance and conservation of fruits and vegetables concerning investment in refrigerated facilities and structures. Our study highlights the relevance of the partnership of retail and wholesale companies with associations that aim at waste reduction. From contributions to the public policy formation, this research points out flaws in the integrated management of the chain, about the adoption of collaborative partnership relationships, information sharing, and lasting partnerships. It leads to several opportunities for creating policies that aim to educate the end consumer about proper handling, planning domestic consumption, making full use of food and conservation, and creating tax incentives to expand the cold chain along the production chain for perishable foods. In general, we expect that this research's results will encourage the adoption of good practices that reduce food waste as a whole minimizing economic, social, and environmental impacts.

Furthermore, we should point out some limitations of the research. The first was the failure to use the same protocol for selecting and filtering papers in the national and international systematic literature reviews. The Methodi Ordinatio, proposed by Pagani et al. [46], was used only in the international review since national journals did not have the impact factor. Regarding the empirical phase of the research, related to case studies, the type of retailer studied was limited to grocery stores and fruit shops. The type of food studied covered only the category of fruits and vegetables, which excludes all the other types of food waste. We also considered the food waste generated in the logistics processes and not that after consumption.

Considering the limitations, some suggestions for future research are proposed. The first is to conduct a more comprehensive systematic literature review, following a harmonized protocol. It would also be interesting to study food waste in other retail links, such as supermarkets of different sizes, restaurants, and after the consumption phase. Further studies can focus on other types of food waste, such as non-perishable and industrialized foods. Additionally, future studies may cover the entire agri-food supply chain, including the logistics processes occurring in the suppliers and the production, the waste in the consumption stage, and the final consumers' role.

Further studies can approach the modeling approach from operational research to propose, for example, a volunteer-based crowd-shipping program for food rescue, considering the restaurant's food donation delivery as proposed by Mittal et al. [49]. Similar modeling can be conducted in the case of wholesaler donations. In addition to that, other models can use modeling from the multicriteria decision aid approach to optimize food donation, helping food-insecure people.

**Author Contributions:** Conceptualization, P.G. and R.C.C.d.A.; Methodology, P.G. and R.C.C.d.A.; Formal Analysis, P.G. and R.C.C.d.A.; Investigation, P.G. and R.C.C.d.A.; Writing—Original Draft Preparation, P.G., K.M.T. and E.A.d.M.W.; Writing—Review & Editing, P.G., K.M.T. and E.A.d.M.W.; Visualization, P.G., K.M.T. and E.A.d.M.W.; Supervision, P.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** We acknowledge the Brazilian Council for the Improvement of Higher Education (CAPES) for its support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Transparency in Global Agribusiness: Transforming Brazil's Soybean Supply Chain Based on Companies' Accountability**

**Gabriel Medina \* and Karim Thomé**

Faculty of Agronomy and Veterinary Medicine, University of Brasilia, Brasília 70910-900, Brazil; thome@unb.br **\*** Correspondence: gabriel.medina@unb.br

**Abstract:** *Background:* Although agri-food supply chains have become fundamental for food security throughout the world, some are associated with negative environmental and socioeconomic impacts. This study explores the possibilities of transforming the governance in Brazil's soybean supply chain based on stakeholders' accountability. *Methods:* We used secondary data from companies' reports and statistical yearbooks to identify key stakeholders in the soybean supply chain as well as to explore trade-offs between reducing farming expansion into new agricultural frontiers and increasing investments in agro-industrial sectors. *Results:* The results reveal that at the global level, multinational corporations along with domestic groups should be held accountable for improving the governance of the soybean supply chain in Brazil since foreign multinationals control 65.4% of it. At the domestic level, losses in Brazil's farming sector can either be offset by an 11% or 5.2% market share increase in the trading segment or in the whole supply chain, respectively, since Brazilian groups control 93.4% of the farming sector but only 7.1% of the agro-industrial sectors. *Conclusions:* Global accountability and domestic trade-offs are fundamental for transforming governance in global agri-food supply chains. They serve as a means for overcoming the current strategy of expansion into new farming frontiers.

**Keywords:** food supply chains transformation; stakeholder accountability; business evolution; corporate environmental management; responses to environmental issues; environmental; social and governance values (ESG)

#### **1. Introduction**

There is growing global interest in the transparency and sustainability of agri-food supply chains [1,2]. Improved relationship strategies in food supply chains (e.g., cooperation, coordination, and collaboration and accountability) can generate positive effects such sustainable gains in environmental and economic dimensions [3].

As some agri-food supply chains are largely international, making them more accountable and sustainable requires a collaborative effort among different countries and stakeholders [4]. Approaches toward sustainable and responsible agri-food supply chains, therefore, need to be promoted in different business sectors that include all the key stakeholders established along the supply chain [5–7].

There is a growing demand for transparency at the international market level over how supply chains source agricultural commodities [8]. As a practical outcome of flaws in transparency, part of the agricultural global flows remains unaccountable, which is detrimental to the agri-food sector as a whole.

Corporate sustainability as a business response to environmental issues requires improved transparency, governance, and accountability across the supply chain as a whole [7]. The existing top-down approaches such as market moratoriums [9], reductions in foreign direct investments [10], and commitments to eliminate deforestation from agricultural commodity chains [11] put pressure on domestic and multinational groups operating in the farming and trading sectors. However, by focusing on output segments, these approaches

**Citation:** Medina, G.; Thomé, K. Transparency in Global Agribusiness: Transforming Brazil's Soybean Supply Chain Based on Companies' Accountability. *Logistics* **2021**, *5*, 58. https://doi.org/10.3390/logistics 5030058

Academic Editor: Robert Handfield

Received: 28 July 2021 Accepted: 20 August 2021 Published: 25 August 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/).

do not hold accountable the important multinational corporations involved in the input segments, such as seeds, machinery, agrochemicals, and fertilisers [8,12].

These approaches also offer no trade-offs between reducing domestic investments in farming expansion into new agricultural frontiers and increasing them in agro-industrial sectors such as seeds, machinery, fertilisers, agrochemicals, and trading. The opportunity costs of reducing the expansion of agricultural frontiers can be offset by investments in industrial segments that better remunerate capital and labour [12]. As a consequence, important stakeholders lack accountability in these current approaches to improve governance [13].

To improve the accountability in supply chains, those who will be sharing the costs for better practices, and the possible trade-offs for committed stakeholders, must be identified. There are two fundamental steps required to make this happen. First, the key stakeholders that need to be accountable in the process must be identified for the sake of transparency. Second, possible win-win solutions and trade-offs for addressing the existing challenges should be explored [14]. To this end, it is fundamental to have a comprehensive understanding of agri-food supply chains as a whole, as well as the market share held by key stakeholders.

This article adds to existing efforts to improve transparency in the soybean supply chain established in Brazil as a means to transform governance based on the accountability of companies and possible trade-offs for committed stakeholders. This can lead to future renegotiations of responsibilities targeting the development of fair, responsible, and sustainable agri-food supply chains.

This article specifically aims to identify the key stakeholders operating in the supply chain and their respective market shares as a means to discuss their accountability in the business. Further, it explores trade-offs between reducing the domestic investments in farming expansion and increasing the market share in agro-industrial segments that better remunerate capital and labour. By doing so, this study explores a novel conceptualisation of food supply chain transformations that can lead to greater benefits for different countries and stakeholders.

#### **2. Theoretical Framework**

#### *2.1. Soybean Agribusiness*

Soybean is the main crop in Brazil, both in scale and in value. In the 1990s, soybean advanced from the south towards the central area of Brazil, and in the 2000s, it expanded farther to the north [15]. Soybean monoculture is now expanding towards new agricultural frontiers such as parts of the Amazon and the Matopiba (Matopiba is a new agricultural frontier in Brazil that partially covers the states of Maranhão, Tocantins, Piauí, and Bahia) in the north and northeast of Brazil, respectively, as represented in Figure 1.

Brazilian environmental law and the forest code disciplines agricultural expansion into areas of native forests but the growing markets for soybeans drive farmers' expansion towards new agricultural frontiers [16]. Part of the commodity-exporting farming sector strategy has been to laterally expand production into these new agricultural frontiers, particularly by reducing the so-called "Brazil cost" (transaction costs of producing and commercialising in Brazil), through the improvement of infrastructure (i.e., roads, storage conditions, railway construction, and improvements in ports) [17].

The most common implications from the expansion of the commodity-exporting farm sector in Brazil are land related conflicts [18] and the increase in deforestation rates [19]. Most commonly, land related conflicts happen when the local population, who are considered informal land holders (the so-called posseiros) are faced with immigrating large-scale farmers in search of new land to be farmed. There here are also similar cases of largescale farmers invading traditional indigenous lands that are still not acknowledged by the government [20].

**Figure 1.** Consolidation and expansion of soy plantations in Brazil between 2006 and 2017. Consolidation is defined as municipalities that have remained as important producers and that have increased their planted areas, when comparing data from 2017 with 2006. Expansion is defined as municipalities planted with soy in 2017 that did not previously have soy planted in 2006. Source: Based on the data of the area planted with soybeans in 2006 and 2017, which are available at the Brazilian Institute for Geography and Statistics website (SIDRA/IBGE). 2017 is the year of the latest agricultural census in Brazil.

The expansion of the agricultural frontier has been causing relevant environmental impacts, particularly the deforestation of native forests. Farming expansion is one of the main causes of deforestation in the Amazon [21] and soybean expansion into the Matopiba region is taking place in the states with the greatest percentage of native vegetation of the Cerrado biome [17].

The growing demand for transparency at the international market level over how supply chains source agricultural commodities has resulted in important practical measures for curbing deforestation. Initiatives led by importing countries include the Amsterdam Declaration, a commitment by seven European countries to eliminate deforestation from agricultural commodity chains [11]. Brazil's Soy Moratorium was the first voluntary zerodeforestation agreement implemented by major soybean traders agreeing not to purchase

soy grown on deforested lands [9]. Recently, these efforts have also been underpinned by investors threatening to withdraw investments from companies connected to deforestation in the Amazon [10].

Initiatives toward sustainable supply chains focus on the traceability of goods produced by stakeholders established at chains (mainly farmers, soybean crushing plants, and traders). These stakeholders try to collaborate with members that are willing to adapt their practices to new market demands toward a free deforestation supply chain, as described in [7]. Even though progress has been made through these efforts, non-committed sectors maintain their business-as-usual practices by accessing less restrictive markets or leaking farming activities to other agricultural frontiers, such as the Cerrado Brazilian Savannah [11,16,21].

While only six large soy traders account for most soy exports from Brazil, soy businesses involve thousands of companies who may have different deforestation footprints, and must independently comply with any voluntary moratoriums [22]. Although most of Brazil's agricultural output is deforestation-free, 2% of properties in the Amazon and Cerrado are responsible for 62% of all potentially illegal deforestation and roughly 20% of soy exports from both biomes to the EU may be contaminated with illegal deforestation [23]. These studies reveal the importance of targeting key market leading companies but also the need for considering the whole supply chain as a means for improving transparency and accountability [22,23].

#### *2.2. Agribusiness Companies' Accountability*

Since the advent of the theoretical framework about the organisation–environment interface demonstrating that no business is an island, scholars have developed intersectoral analytical approaches involving whole supply chains [24]. Specifically in agri-food, supply chains include the farming sector, and the total sum of operations involved in the manufacture and distribution of farm supplies; the production operations of the farm; and the storage, processing, and distribution of farm commodities made from them [25].

Thus, firms are seen as no longer competing only as independent units, but also as members of supply chains or networks, connected to one another to organise and provide a product or service [26]. In this perspective, agri-food supply chains developed the ability to integrate themselves into a chain as a strategy to succeed, creating internal governances in order to increase coopetition [3]. Several challenges and opportunities arise from taking part in a supply chain that forces firms to look for more effective forms of coordinating flows, inside and outside the focal firm [27].

Considering agricultural and agro-industrial practices, new concerns have emerged [7], specifically in areas of the environment [12], transparency [8], and sustainability [28]. Therefore, the development of sustainable food supply chains has gained prominence in recent years [5], with special attention for local characteristics [1] and ecological, environmental, and social dimensions [3].

Through a food supply chain, stakeholders are characterised by interdependence [29]. Scholars sustain that beyond the interdependence based on economic behaviour and through the convergence of goals and needs, the stakeholders' interaction and behaviour do not represent a disconnection from the scenario of social, cultural, and environmental reality, but are also situated on those [3]. The supply chain approach compose new frames that demand a new governance arrangement [5], and in some cases, a redesign of the food supply chain [30].

New frames create specific demands for specific resources that do not just address economic integration, but also correspond to scenarios that involve ecological, environmental, and social dimensions [29]. It is the case for globalised supply chains [5,31], and with multi-sectorial partnerships, as in the soybean produced in Brazil [8,12].

Thus, actions need to be coordinated and accountable to achieve a goal and determine responsibilities among supply chain stakeholders [7], [31]. A supply chain governance framework helps to better understand managerial issues such as how to select partners, how to go about designing partnership arrangements, and how partnerships can be developed to ensure long-term supply chain sustainability and success [32].

Stakeholders in supply chains create their own internal governance arrangements and a variety of external stakeholders may also seek to influence chain activities and/or outcomes [5]. This means that supply chain governance has a multi-institutional nature [5], besides being multi-sectorial [8,12]. Building on this background, studies on supply chains' governance reinforce the importance of strategic alliances among stakeholders [33] realised in socio-technical arrangements [34].

This comprehensive approach helps us to address the functioning of agri-food supply chains, the role played by key stakeholders, and by exploring improvements in its governance based on stakeholders' accountability and trade-offs. It is also useful for identifying the responsibilities and alliances of different stakeholders particularly in hierarchical structures. Even though it is focused on private firms, it can be complemented by a more comprehensive approach involving other stakeholders such as the state.

#### **3. Methods**

Similar to what was conducted in previous studies [8,12], key stakeholders in the soybean supply chain were identified based on the definition of the most popular inputs used in each production stage, their suppliers, and the country of origin of the companies involved. The relevance of each stakeholder was estimated based on their market share in each business segment as described in [35]. Market share information was obtained in the companies' reports as well as in statistical yearbooks from producer associations that annually estimate the market participation of their members in Brazil, as described in Table 1.

**Table 1.** Segments and sources.


For market share estimations, we first quantified the total sales in the country for each input per segment (e.g., 5580 soybean combines sold in Brazil in the 2019/20 agricultural year). We then identified the main international and domestic companies operating in each segment (e.g., CNH, John Deere, and AGCO in the case of soybean combines), and their total sales (e.g., 2903 soybean combines by CNH, 2269 by John Deere, and 408 by AGCO). To estimate the total market share of domestic groups in each segment of the supply chains in the sample, we calculated and added the market shares of all Brazilian groups. The results are found in Table 2. Information was collected for the years 2015 and 2020 as a means to measure market share evolution over time.

Trade-offs between reduced domestic investments in the farming sector and the increased market share of domestic groups in the agro-industrial segments were explored. This was performed according to the market size of each soybean supply chain stage, which was estimated based on the literature review. The market size of all studied segments (from seeds to trading) was estimated based on annual market transactions in each segment, as presented in Table 3. From this information, scenarios were projected to offset the opportunity costs of curbing agricultural expansion into the two main agricultural frontiers in Brazil: The Amazon with 4.5 million hectares planted with soybean and the Matopiba region of the Cerrado with 5.7 million hectares planted with soybean.

**Table 2.** Market share held by Brazilian vis à vis multinational companies in sectors of the soybean supply chain established in Brazil for the years of 2015 and 2020 (in percentage).


Source: Based on data published by Anprosem [36], Anda [37], Aenda [38], Anfavea [39] and Aprosoja [40].

Trade-offs between reducing investments in farming expansion and increasing market share in agro-industrial sector were explored for the following scenarios: increased domestic market share in the trading segment and in the whole supply chain. Highlighting the significance of agri-food in 2019, the agri-food sector was responsible for 21.2 % of the

Brazilian gross domestic product (GDP), while the farming sector represented 4.8% of the national GDP [41].


**Table 3.** Market size of each business segments (in USD billions).

\* Based on raw materials and not on manufactured fertilisers.

#### **4. Results**

#### *4.1. Key Stakeholders Based on Market Shares*

In Brazil, 91.8% of the soybean cultivated is transgenic and the German multinational Bayer controls 90% of Brazil's transgenic market share. Although Brazil has companies that dominate soy genetics, transgenics are controlled by multinationals that receive royalties from Brazilian companies licensed to use their technology in seed production. Domestic seed producers such as Tropical Melhoramento and Genética (TMG) who have created their own germplasm improvement programmes and pay royalties for the use of transgenics, hold 25% of the market share. Studies show that multinationals that own the characteristics transferred to local germplasm make about 65% of the profit from the final price of soybeans, while the other 35% of the profit is shared between the germplasm developers and seed multipliers [45]. Thus, in the segment of the chain related to the production of seeds, domestic capital would be equivalent to only 8.7% for the agricultural year of 2019/2020 (35% of 25% market share), as described in Table 2.

The machinery sector is a worldwide oligopoly as a result of mergers and acquisitions headed by the following major international groups: John Deere, CNH (holder of the brands Case and New Holland), and AGCO (holder of the brands Massey Ferguson and Valtra). In Brazil, the three companies together control 99.6% of tractor sales and 100% of combine harvester sales [39]. The national capital share for the agricultural year of 2019/2020 was estimated at 0.2% when including the Brazilian company Agrale, see Table 2, which produces tractors rarely used for soybean due to their relatively small size. There is a greater market share of domestic companies in the case of agricultural implements, such as ploughs, scarifiers, limestone spreaders and cultivators, although precise data on market share is not available.

The following two types of companies operate in the fertiliser segment: those that produce raw materials (or simple fertilisers) and those that manufacture formulated fertilisers. Most of the raw material for the fertilisers used in Brazil is imported. In the case of soybeans, phosphorus (44% imported) and potassium (95% imported) are the most commonly used macronutrients [37], since soybeans do not require nitrogen fertilisation and there is little use of micronutrient fertilisation. In Brazil, the Vale Company, controlled by Brazilian groups, used to be the largest producer of phosphorus and the only producer of potassium, a sector that is now controlled by the multinational Mosaic. It is estimated that domestic groups produce 8.7% of the fertilisers consumed in Brazil (17.5% of phosphorus and 0% of potassium), as in Table 2.

In relation to fertiliser manufacturers, the market in Brazil is led by the multinational Yara, with domestic groups holding 29.8% of the market. The Fertipar Group and Heringer (today with 56% of national capital) are the Brazilian companies with the largest participation in the manufacturing of fertilisers in Brazil. The rest of the market is serviced by domestic companies of a regional nature and by multinational groups. Considering an 8.7% national share in the production of raw materials and a 29.8% share in the production of fertilisers, it is estimated that Brazilian participation in the fertiliser market has an average of 19.2%, as shown in Table 2.

In Brazil, 94% of total agrochemical (pesticides) sales refer to the following three classes of products, defined by their purpose: insecticides (33%), herbicides (32%), and fungicides (29%). Soybean farming is the main consumer of agrochemicals in Brazil, accounting for 50% of sales according to the National Union of the Plant Defence Products Industry [46]. There are the following two business segments: products with patents that require innovation, controlled by multinational groups; and generic products, authorised after patent exclusivity periods end, in which the industry with domestic capital still has a stake. In the segment of products with patents, there is ample competition, but few of them have a significant market share. In Brazil, the multinationals control 94.2% of sales, specifically Syngenta/ChemChina (18.6%), Bayer (15.7%), and Basf (9.2%), and other multinational groups with smaller slices. The companies with national capital only make up 5.8% of the total of commercial agrochemicals traded in the country [38]. This percentage is made up of domestic companies such as Nortox and Ourofino, and a group of small businesses.

Brazil has been experiencing changes in the profile of the soybean grower. The private producer is now competing with large national corporations and multinational companies such as Los Grobo, which leases land and manages crops, and Agrinvest, which purchases land for agricultural production. In Brazil, there are 33,200 registered properties belonging to foreigners, occupying 3.8 million hectares [47]. The area used for soybeans in Brazil is around 57.2 million hectares and it has been estimated that 93.4% of this area belongs to Brazilian farmers (see Table 2).

The large multinational export companies such as ADM, Bunge, Cargill, and Dreyfus (known as the ABCD group) have oligopolised the governance of the soybean supply chain [48]. Recently, China celebrated the purchase of Noble Agri (trade) by China National Cereals, Oils, and Foodstuffs Corporation (COFCO) as a way to ensure their presence in 21 countries, including Brazil and Argentina, its two largest soy suppliers. It is estimated that domestic capital controlled about 16.1% of the commercial soybean market in the country in the agricultural year of 2019/2020, less than the 30.7% market share of 2015, as in Table 2. Brazilian groups include Amaggi, Coamo (a cooperative), Cutrale, Bianchini, Granol, Caramuru, and Comigo (a cooperative).

The results reveal that the market share held by Brazilian groups as a whole dropped between 2015 and 2020. In this time span, there were changes in the share held by domestic companies in the segments of seeds (from 16.5 to 8.7%), fertilisers (from 33.5 to 19.2%), pesticides (from 4.3 to 5.8%), machinery (from 1.9 to 0.2%), and in the trading sector (from 30.7 to 16.1%). Proportionally, the market share held by Brazilian groups as a whole dropped from 40% in 2015 to 34.6% in 2019/2020. The share of domestic groups in the capital and technology intensive sectors (excluding the farming sector) dropped from 12.5% in 2015 to 7.1% in 2020.

Figure 2 summarises the market share held by Brazilian vis à vis multinational companies in key sectors of Brazil's soybean supply chain for the agricultural year of 2019/2020. It highlights that key agro-industrial segments (such as seeds, machinery, and pesticides) are controlled mainly by multinational companies.

**Figure 2.** Market share held by Brazilian vis à vis multinational companies in key production stage of the soybean supply chain established in Brazil in 2020 (in %). Source: Based on data published by Anprosem [36], Anda [37], Aenda [38], Anfavea [39] and Aprosoja [40].).

While domestic groups have an important market share in the farming sector, multinational companies tend to control the industrial segments of the production chain. German companies control the seed sector and have an important market share in the pesticides sector. American companies control the machinery sector and have important shares of the pesticides and trading sectors. Chinese companies have a relevant market share in the pesticide and trading sectors, as described in Figure 3. The soybean produced in Brazil is exported mainly to China (57.4%) and domestic consumption in Brazil is the second largest market (15.6%) [48].

**Figure 3.** Home countries of companies controlling key segments of Brazil's soybean supply chain by 2020. Source: Based on data published by Anprosem [36], Anda [37], Aenda [38], Anfavea [39] and Aprosoja [40].

#### *4.2. Trade-Offs Based on Market Size*

The studied segments of the soybean production chain in Brazil generated USD 86.9 billion worth of gross income in the agricultural year of 2019/2020 (see Table 3). The seed segment had a USD 2 billion income, while sales of tractors and combine harvesters resulted in a USD 2.6 billion gross revenue, as described by companies' reports. Fertilisers generated a USD 4.7 billion gross income and the companies in the agrochemicals segment had a gross income of USD 8.1 billion. Estimations were also made for the farming and trading segments that accounted for USD 28.6 billion and USD 41.2 billion worth of markets, respectively [44].

Figure 4 summarises the market size of each segment in the soybean production chain established in Brazil and the market share held by domestic vis à vis multinational companies. It also presents two possible scenarios for trade-offs between reduced farming expansion into new agricultural frontiers and increased market share in agro-industrial sectors. In the first scenario, the opportunity costs of hindered farming expansion are offset by an 11% domestic market share increase in the trading segment. In the second scenario, the opportunity costs are offset by a 5.2% domestic market share increase in the whole supply chain.

**Figure 4.** Market size of key segments of the soybean supply chain for the agricultural year of 2019/2020 and possible financial trade-offs between farming expansion and increasing market share in industrial segments (in USD billions). Source: Based on data published by Anprosem [36], Anda [37], Santos and Glass [49] and Escher and Wilkinson [44].

Both scenarios estimate the opportunity costs of stopping the two main agricultural frontiers in Brazil: the Amazon frontier with 4.5 million hectares planted with soybean and the Matopiba region of the Brazilian Savannah (Cerrado) with 5.7 million hectares. Soybean farming generates USD 3.5 billion income per year in the Amazon and USD 4.5 billion per year in the Matopiba, as shown in Table 3.

The results reveal that a USD 4.5 billion gross income in farming activities can either be offset by an 11 or 5.2% domestic market share increase in the trading segment or in the whole supply chain, respectively, as shown in Table 4. In the first scenario, a domestic share in the trading segment equivalent to 27% of the total (the current 16% plus 11%) would be enough for offsetting the income of soybean farms in the Amazon or in the Matopiba region (USD 4.5 billion). In the second scenario, the same outcome can be obtained with a 12.3% domestic market share in the whole soybean production chain (the current 7.1% plus 5.2%). In both cases, the estimated increased market share is smaller than the one held by domestic groups in 2015, that is, 30.7% in the trading segment and 12.5% in the industrial segments.

**Current Situation Projection Outcome** Expansion into agricultural frontiers Area with soybeans in Brazil (million ha) Gross income (USD billion) Measure (Curb deforestation) Area with soybeans (million ha) % Income (USD billion) Amazon 36.4 28.6 100% 4.5 12.4 3.5 Matopiba 36.4 28.6 100% 5.7 15.7 4.5 Increased market share Current domestic share Gross income generated (USD billion) Measure (increase in domestic market share by) Targeted market share % Income generated (USD billion) Trading 16.0 41.2 11% 27.0 6.6 4.5

**Table 4.** Trade-offs between reducing investments in farming expansion and increasing market share in agro-industrial segments.

Whole chain 7.1 86.9 5.2% 12.3 6.2 4.5 Source: Based on data published by Anprosem [36], Anda [37], Aenda [38], Anfavea [39] and Aprosoja [40].

#### **5. Discussion**

Efforts to promote sustainable and responsible agri-food supply chains have focused on top-down approaches such as market moratoriums [9], reductions in foreign direct investments [10], and commitments to eliminate deforestation from agricultural commodity chains [11]. These measures have proved efficacious in committing the more modern agrifood supply chain to initiatives such as the soy moratorium [9]. However, current efforts still need to address the following two important issues for improved transparency and governance: 1. Make all the stakeholders accountable, including multinational companies and 2. Explore the trade-offs between reduced farming expansion into new agricultural frontiers and domestic investments in agro-industrial segments.

Sectors not committed to improved governance tend to maintain their business-asusual practices and undermine the whole supply chain reputation [11]. Therefore, topdown enforcement approaches have to be complemented by horizontal efforts to improve accountability in agri-food supply chains. To this end, it is fundamental that supply chains are well understood, and that the roles and responsibilities of key stakeholders are made transparent and, in some cases, renegotiated [50]. It is possible based on a broad picture of the supply chains including the market share held by the key stakeholders.

Corporate sustainability in agri-food supply chains requires improved accountability across the supply chain as a whole [7]. Agri-food supply chains include the farming sector, and the total sum of operations involved in the manufacture and distribution of farm supplies as well as the processing and distribution of farm commodities [3]. Several opportunities, but also responsibilities, arise from taking part in a supply chain [5]. New concerns have emerged in agri-food supply chains [7], specifically in areas of the environment [12], transparency [8], and sustainability [28]. For addressing these issues, recent studies reveal that the supply chains' governance should be based on multi-stakeholder efforts [5,8,12].

This study reveals that multinational corporations are the key stakeholders in the soybean production chain established in Brazil. In all the soybean supply chain agroindustrial segments, multinationals hold the majority of the market share and this foreign control has grown in recent years [49]. Given the hierarchies established in the hybrid forms of governance in supply chains [12], powerful stakeholders such as multinational corporations can play a role in setting up strategic alliances for improved governance [28].

This study also adds to current efforts focusing on the output sectors such as farmers [21] and traders [48] by providing a comprehensive understanding of the whole supply chain including the input sectors (seeds, fertilisers, agrochemicals, and machinery) that should also take part in an improved effort for accountability (see Figure 2). Seeds and pesticides companies have been developing new varieties adapted to new agricultural frontiers prompting conflicts among farmers and among farmers and local communities [51]. Tractors and other machinery are used to introduce plantations in illegally deforested areas [52], and fertilisers make planting soybean in new agricultural frontiers viable [53]. Ultimately, these are business sectors profiting from the expansion of agricultural frontiers [54].

Although the governance is part of companies' strategies for economic performance, in the soybean supply chain case, the existing governance has not avoided negative social and environmental externalities [17]. Therefore, besides the private sector, governments also have a role to play in promoting transparency and supporting sustainable business. As most of the multinational companies are based in wealthy economy countries such as the United States, Germany, and China, these countries should take an active role in promoting the accountability of their home companies operating in developing countries [55].

Local stakeholders including companies, governments, farmers, and local communities can also play a role in improved governance [1]. Domestic groups can explore trade-offs between reducing farming expansion and increasing their market share throughout agroindustrial segments upstream and downstream of farms, as represented in Figure 4. The advances of agri-food production in Brazil offer areas of opportunities for Brazilian groups ranging from strengthening domestic seed-producing companies to the consolidation of regional trading companies. It is by investing in the agro-industrial sectors that will better remunerate capital and labour [56], and going beyond the current focus on the primary production of commodities, that developing countries will benefit from agri-food expansion for their development [57].

To this end, agricultural policy in Brazil needs to evolve from the current almost exclusive focus on subsidised credit (mainly funding or "working" credit for large farmers) to more comprehensive investments that can bring longer-term returns to the agri-food sector as a whole [17]). Rural credit implies the risk of serving only to compensate the low profit margins of farmers who operate in the agri-food supply chains increasingly controlled by multinational groups. As it is now, credits are used by farmers as working capital to buy inputs such as seeds and pesticides from multinationals, which ultimately implies transferring money from Brazilian taxpayers to foreign corporations.

#### **6. Conclusions**

This study adds to existing efforts to improve transparency in the soybean supply chain by revealing the key companies operating in each production stage (from seeds to trading) as a means to transform the business based on companies' accountability. The results highlight that all the firms established along the supply chain should be held accountable for improved governance. The efforts for doing so should involve both multinational corporations and domestic groups.

Specifically, this study reveals that multinational corporations established along Brazil's soybean supply chain controls 91.3% of the seed, 99.8% of the machinery, 80.8% of the fertiliser, 94.2% of the agrochemicals, and 83.9% of the trading sectors. German companies control the seeds sector and have an important market share in the agrochemical business. Companies in the US control the machinery sector and have important shares in the agrochemical and trading sectors. Chinese groups have a relevant market share in the agrochemical and trading sectors.

As Brazilian groups control 93.4% of the farming sector, but only 7.1% of the agroindustrial sector, they should explore trade-offs between reducing investments in farming expansion into new agricultural frontiers and increasing their market share in agroindustrial segments. This study reveals two possible trade-offs between reduced farming expansion and increasing the industrial market share. The opportunity costs of hindered farming expansion can be offset either by an 11% domestic market share growth in the trading segment or by a 5.2% domestic market share growth in the whole supply chain. Domestic investments in the agro-industrial segments should be promoted as an alternative to investments in farming expansion into new agricultural frontiers.

These results reveal the need for structural changes as a means to improve governance and corporate sustainability in the soybean supply chain. The necessary improvements include i) increasing the accountability of multinational companies once they are the gearing parts of the soybean agri-food supply chain in Brazil and elsewhere; ii) increasing the market share held by domestic agro-industrial companies in domestic business as a means to promote regional development; iii) discouraging farmer expansion into new agricultural frontiers with high social and environmental costs by replacing the current agricultural policy focused on credit for large farmers through comprehensive investments targeting agri-food supply chains as a whole. By unveiling it, this study highlights the need for a novel conceptualisation of food supply chain transformations that can lead to greater benefits for different countries and stakeholders.

**Author Contributions:** Conceptualisation, G.M.; methodology, G.M.; investigation, G.M.; data curation, G.M.; analysis and draft preparation, G.M.; writing—review and editing, K.T.; literature review, K.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Dada supporting reported results can be found at: http://dx.doi.org/ 10.5801/ncn.v24i1.8521.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Review* **Food Supply Chain Transformation through Technology and Future Research Directions—A Systematic Review**

**Ahmed Zainul Abideen 1, Veera Pandiyan Kaliani Sundram 1,2,\*, Jaafar Pyeman 1,2,\*, Abdul Kadir Othman 1,2 and Shahryar Sorooshian 3,4**


**Abstract:** *Background*: Digital and smart supply chains are reforming the food chain to help eliminate waste, improve food safety, and reduce the possibility of a global food catastrophe. The globe currently faces numerous food-related issues, ranging from a lack of biodiversity to excessive waste, and from ill health caused by excessive consumption to widespread food insecurity. It is time to look back at how technology has tackled food supply-chain challenges related to quality, safety, and sustainability over the last decade. Moreover, continuous transformations of the food supply chain into a more sustainable business model with utmost resilience is the need of the hour due to COVID-19 disruptions. *Method*: This study aimed to systematize literature (2010–2021) in the described context and propose a future research direction, with the assistance of a systematic review and bibliometric analysis on the research agenda proposed above. *Results*: The findings reveal that technological Industry 4.0 (IR 4.0) tools face specific barriers due to the scope and objective of the application. *Conclusion*: The Internet of Things has received more attention than any other IR 4.0 tool. More integration between the specialized tools is needed to address this issue. Furthermore, the authors have proposed a food supply chain-based operational framework on technological inclusion to facilitate the roadmap for food supply chain 4.0 for more resilience and food supply chain viability.

**Keywords:** food supply transformation; supply chain 4.0; food safety; food quality; food sustainability; COVID-19 disruptions; systematic review

#### **1. Introduction**

The need for food is indicated to quadruple over the next ten years, and the only acceptable alternative is to increase supply without jeopardizing our future. According to the most current UN estimate, there are 7.3 billion people today—and we may reach 9.7 billion by 2050. This expansion, together with rising affluence in developing nations (which generate dietary changes such as eating more protein and meat), is pushing increased global food demand. By 2050, food demand is anticipated to increase by 59 percent to 98 percent. This will shape agricultural markets in unprecedented ways. Farmers worldwide will need to enhance crop production, either by increasing crop production on existing agricultural land or by raising crop productivity on existing agricultural lands through fertilizer and irrigation, as well as adopting innovative methods such as precision farming. However, the environmental and social costs of clearing more land for agriculture are often significant, especially in the tropics. Moreover, crop yields (the number of crops gathered per unit of area cultivated) are currently expanding too slowly to satisfy projected food demand [1]. As a result, farmers' adoption of technology is a critical method for improving

**Citation:** Abideen, A.Z.; Sundram, V.P.K.; Pyeman, J.; Othman, A.K.; Sorooshian, S. Food Supply Chain Transformation through Technology and Future Research Directions—A Systematic Review. *Logistics* **2021**, *5*, 83. https://doi.org/10.3390/ logistics5040083

Academic Editors: Karim Marini Thomé, Michael Bourlakis and Patricia Guarnieri

Received: 28 October 2021 Accepted: 22 November 2021 Published: 25 November 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/).

agricultural sustainability and production in developing countries [2]. Farm technology, such as remote-controlled harvesting, automated irrigation systems, biometric scanners, drone-based inventory monitoring, and driverless tractors, has made a big difference in recent years. However, the agriculture industry is not as digitally advanced as other industries [3]. Technology can help farmers improve transparency and traceability along their supply chains. Consumers have acquired access to sustainability- and compliance-related information because they are now keen on tracing and tracking the food source they consume [4]. This has further pushed all the stakeholders in the food supply chain (FSC) to create a strong connection between sustainable practices and the food value chain [5].

The Sustainable Development Goals are centered on food systems. The SDGs' broad scope necessitates holistic methodologies that include previously "siloed" food sustainability analyses [6]. All components of food systems must be sustainable, resilient, and efficient in order to provide food and nutrition security for current and future generations. To promote food system sustainability transitions, several measures can be undertaken, including increased efficiency, demand limitation, and food system change. Creating sustainable food systems necessitates shifting from a conventional agriculture-centered policy to a smart food system policy and research paradigm [7]. Sustainability and environmental protection have been in the spotlight. Sustainability is having a significant impact on the global food supply chain, partly because customers desire healthier foods that do not harm the environment [8,9]. Technological tools such as artificial intelligence (AI), Machine Learning (ML), Internet of Things (IoT), Big Data (BD), Digital Twins (DT), Blockchain (BC), and Cyber-Physical Systems (CPS) have leveraged their capabilities greatly to address food supply-chain challenges related to safety, quality, traceability, and sustainability. There is a need to systematize past research endeavors to understand better the trends and future research scope in this context. On that note, this research aimed at conducting an integrated approach of a systematic review and bibliometric analysis that focused on answering the following research questions: What are the current challenges in FSC? What are the technological applications in FSC to overcome those challenges especially during pandemic disruptions? Why is sustainable FSC so important for the future? What are the antecedents of effective relationship management and FSC transformation?

The introduction part of this paper discussed the study's objective, and is followed by the literature review that portrays the trends, applications, and benefits of different technological tools applied in FSC. The keyword selection and article exclusion/inclusion criteria are described in the methodology section, followed by the results section. The dataset was snowballed with systematic and bibliometric analysis to assess the research trend and gaps. Using the insights accumulated from the overall review, the authors have proposed future research directions and barriers in the technological adoption in FSC at the end of this paper before the conclusion.

#### **2. Literature Review**

#### *2.1. Rubrics of Food Supply Chain*

Food production is divided into four phases. The first stage is locating (local or international) raw materials and verifying their quality and safety standards. Next, after the food is processed, it is sent to the handling and storage stage, where it is cleaned and processed into various end products. The subsequent phase comprises handling and storage, where they are packed according to their specifications before being moved on to distribution and transportation [10]. There are different supply-chain models, such as continuous (cash crops), fast chain (perishable items), efficient (unique products), agile (retail products), flexible (agricultural and meat products) and custom figured (hybrid food items).

Moreover, the global food supply chain is complex and struggles to meet the sustainability and safety benchmark. Therefore, a more robust supply chain structure and market governance are needed to maintain an innovative, sustainable food system. Furthermore, sustainability, availability, financial capital, food safety and security, and traceability are crucial to building a smooth FSC [11].

#### *2.2. Effect of Pandemic Disruptions on Food Supply Chain*

The food systems are meeting enormous stress and challenges due to the pandemic disruptions. The world food manufacturers and supply chain providers are now trying to meet that demand by using effective international and domestic trading protocols to stop supply chain resources and bottlenecks [12].

The COVID-19 epidemic has ushered in a new era in the world, with FSC bearing the full brunt. Considering the food supply chain, commercial activities and the supply of various food products have been halted due to a reduction in demand, the closure of food manufacturing facilities, and financial constraints. Farm labor, processing, transportation, and logistics obstacles, as well as significant shifts in demand. The majority of these disruptions are the result of policies implemented to slow the spread of the virus. In the face of these pressures, food supply chains need resilience. Grocery shop shelves are being emptied at a quick pace as stockpiling activity shifts in conjunction with panic buying behavior among customers. Moreover, the greatest threat to food security is not a lack of food, but a lack of consumer access to food [12,13].

Food policymakers are working hard to maintain costs and flows at as minimal a level as possible. The worst-affected section is labor scarcity in food processing and packaging companies, as the industries have been asked to reduce their workforce to stop transmissions. As a result, there are more significant bottlenecks in the FSC [13].

#### *2.3. Conventional Food Supply Chain and Issues*

As the world's population grows, so does the need for more food, demanding a more excellent supply of high-quality commodities. On the supply side, however, there is still concern about the industry's ability to fulfill higher product yields and quality improvements as a result of issues such as climate change, droughts, and agricultural productivity. The global agricultural linkages are intricate because they involve numerous actors at various levels, from those who generate and add value to processed goods to those who sell. When there are several distinct food items, each with its own unique and widely fragmented supply chain, the complexity rises. Consumers are increasingly concerned about responsible food sources and food production [14]. FSC management is more difficult in developing countries because they typically involve small-scale farmers with hardly any market governance and outreach. Adverse effects on food availability are generated because of the hurdles faced by FSC, such as substantial intermediation, diminished profitability, decreased quality, food waste, and loss of revenue [15].

Therefore, major players are now motivated to adopt sustainable methods in their supply chains since they can guarantee a consistent food supply and profitability. However, sustainability has a price and workflow to follow. It is one of the major trump cards that can fetch an organization's competitive advantage as per the natural resource-based view. The parameters of successful sustainability directly reduce wastes and improve environmentally green practices (waste reduction), social responsibility (social wellbeing), and economic viability (improved livelihood) [16,17]. It would be interesting to see how technological tools assist in addressing these challenges in FSC.

#### *2.4. Application of Internet of Things (IoT), Big Data & Blockchain in FSC*

In underdeveloped countries, only a tiny part of the food supply chain will usually be considered for food ecosystem security audits. The accessibility of the ecosystem, access to the supply chain, and utilization of the food chain are three measurement scales generally used to inspect food and ecosystem security. Food supply networks are complex and interconnected, and IoT-based systems can monitor them to capture details on food materials and protect the ecosystem [18]. The Internet of things (IoT) platform can provide product traceability information in the food supply chain, assisting customers, especially during this pandemic disruption where the information available is so vague. By combining IoT and blockchain technologies, FSC can become more transparent and productive by delivering robust and stable information to clients and related stakeholders [19].

At present, pathogenic and parasitic contaminations can move with frozen food packages, according to scientific evidence, especially in the context of the current COVID-19 situation, where traceability is critical in maintaining food quality and safety. To create a tamperproof audit trail to verify parasites and viruses in packed foods in the FSC, IoTbased, tamperproof data sharing with a centralized architecture and blockchain smart contracts can be used [20]. IoTs can efficiently handle seedling procurement and temperature management in the agriculture industry [21]. Ortañez et al. (2020) [21] built an effective and flexible IoT-based coordinating system for boosting the coordinating mechanism in the agriculture food supply chain during natural outbreaks, to stop the issues caused by fake food. Balamurugan et al. (2021) [22] presented a supplier-based, blockchain hyperledger technology to ensure that FSC data is available and traceable, with an unimpaired substantial computational capacity when implemented within the realms of the IoT [23].

Mondal et al. (2019) [24] presented a distributed ledger technology assisted by IoT architecture, and created a transparent food supply chain using a proof-of-object-based authentication system, similar to cryptocurrency's proof-of-work protocol, coupled with an RFID-connected sensor for real-time data acquisition. As a result, establishing a food traceability supply chain is an effective strategy to address the food safety issue. However, the running costs of a standard food traceability supply chain system are substantial [25]. In an environment where economies are growing more competitive, diversified, and complex, customers have now started to expect high quality and traceability. Blockchain-based software platforms have been advocated to improve traceability by increasing transparency within the FSC [26].

Because of rapid technological advancements, key competitive techniques are rapidly changing. The amount of data globally is continuously increasing; every 12 months, the amount of data in the world doubles [27]. Customers now put too much emphasis on food ingredients and nutritional composition. Even while organic foods are nutritious, they need stringent certification procedures. Big data and blockchain can suffice this issue by providing the necessary certification platform [28].

Li et al. (2017), [29] created a prototype tracking tool that allows the use of sensor data and the creation of data-driven pricing decisions in a variety of operational scenarios and product features. Furthermore, in the same context, Ji et al. (2017), [30] previously introduced a Bayesian network approach for predicting market demand that combines sample data and establishes a cause-and-effect relationship between data, as well as a crisp schematic on how large data can be integrated into Bayesian mathematical network optimization to anticipate demand. Moreover, a service-oriented traceability platform (SOTP) used in the packaged foods supply chain allows real-time dynamic data acquisition and processing of packaged foods information, creating a ubiquitous environment in the packaged foods supply chain. This ensures packaged food's life-cycle visibility and traceability from their production, circulation, and consumption [31]. Additionally, the objective of algorithms for tracing contamination sources and locating potentially contaminated food in markets can be achieved [32].

#### *2.5. Blockchain in FSC*

Blockchain is a secure digital ledger that records and validates user transactions that cannot be altered or deleted. These actions are known as blocks, each having its own digital signature and a connection to the previous one. This approach creates a growing list of chronologically arranged encrypted records. Digital currencies or cryptocurrencies are utilized across the supply chain to pay for the quality of assets. Agriculture farmers, distributors, and consumers can pay for selective access, sharing, and authentication of products. The transactions are followed by advanced encryption systems [33,34]. A QR code is placed on food packaging that contains all of the evidence gathered along the supply chain. Consumers may scan the QR code to obtain comprehensive stock traceability, including origin information. Moreover, in global logistics, the distributed ledger technology-based smart contracts (which use the blockchain to execute agreements), and

the smart web (cloud) have all been used to preserve container information so that its partners may receive data on container conditions, such as humidity and temperature [34].

Furthermore, this allows banks to also benefit from the FSC's visibility and lend money to farmers without risk. Buyers will have an easier time verifying whether the seller's statements regarding the food quality are accurate through blockchain smart contracts [35]. This technology makes it easier to decentralize, enhance security, sustain and manipulate supply chains during disruptions [33]. Furthermore, a better cost-control mechanism of the food traceability supply chain-based system is also possible to practice [25,26].

#### *2.6. Artificial Intelligence (AI) and Machine Learning (ML) in FSC*

AI offers many benefits to the food-processing supply chains. Supply chain players will invest in AI if they foresee long-term revenue gains and other benefits [36]. AI can improve the industry's performance in many ways and add to the gross domestic value. These ways include technical feasibility, intelligence, data quality, and accessibility [37]. Additionally, the food supply chain uses vast amounts of energy. This use significantly affects the environment all along the chain. AI-based optimization can help reduce energy consumption by sharing information, minimizing energy use, optimizing truck routes, reducing greenhouse gas, and shrinking the carbon footprint which is very essential during this pandemic and post pandemic era [38]. Recently, researchers have focused on using AI to help protect supply chains from the effects of disruption. This research suggests that AI can help to improve forecasts and thus mitigate the outcomes of disruptions, an aspect of supply-chain risk management [38]. In recent years, supply-chain risk management has received a lot of attention, intending to protect supply chains from disruptions by forecasting their occurrence and mitigating their negative consequences. Therefore, AI has prompted researchers to look into machine-learning techniques and their application in supply-chain risk management [39].

Food quality is a significant aspect that food engineers keep in mind whilst designing a food system. In tea production, Núñez-Carmona et al. (2021) [40] calculated the volatilome of several tea varieties using metal oxide gas-sensor data and machine learning to provide a competitive tool that can project predictive analysis based on time, costs, and contamination. Moreover, food traceability and shelf life are directly proportional. ML assists blockchain platforms in building anticounterfeiting solid technology in FSC, overcoming drawbacks of low levels of traceability, scalability, and data accuracy. Shahbazi et al. (2021) [41] suggested a blockchain- and machine-learning-based food traceability system (BMLFTS) that relied on a fuzzy logic approach that improved perishable food shelf-life management. The BC was used to reduce warehouse and shipment times and thereby improve reliability. Alfian et al. (2020) [42] proposed an IoT-based traceability system that utilized RFID and raspberry pi-based sensors. The RFID reader tracks and traces the merchandise while the raspberry pi is used during storage and travel to record temperature and humidity and forecast future temperatures. Sometimes, the food supply chain involves multiple stakeholders and distributors, which always leads to information asymmetry. To counteract, Mao et al. (2018) [5] designed a blockchain-based credit evaluation system to enhance food supply-chain monitoring and management efficiency through intelligent and innovative Long Short-Term Memory Network contracts (LSTM).

#### *2.7. Digital Twins & Cyber-Physical Systems in FSC*

The adoption of diverse technologies has aided in the advancement of food processing and logistics. To improve insights and optimize designs and processes, more sophisticated numerical tools and software platforms have emerged. The concept of the digital twin was successfully introduced as a valuable tool in the context of industrial digitization [43,44]. The digital twin is a virtual clone of a real-world process, connected to the environment via Big Data tools to analyze the functions of more physical models. This enables us to model and virtually visualize environments and processes risk-free, which is very apt for

the present COVID-19 conditions [45]. The supply chain-based digital twins provide endto-end visibility along with demand charts, levels of inventory, and asset management [46].

Furthermore, cyber-physical systems have evolved as intelligent mechanical entities that help in the smart production and packaging of products. Therefore, it can be easily linked with IoT, AI, and ML for better performance [47]. One good example of a digital twin application in FSC is portrayed by [48]. They created a digital fruit twin based on mechanistic modelling mimicking the thermal behavior of food products (fruit) across the cold chain, and quantified the enzymatically driven, temperature-dependent biochemical breakdown processes. This improves supply networks by documenting and predicting where temperature-dependent food-quality loss happens in each supply chain due to extended refrigeration times.

#### **3. Methodology**

The selection of keywords and the database were the first steps in this study. The authors used the Scopus database for this study because it enabled them to investigate a broad spectrum of publications. The primary keyword, food supply chain, was entered into a title search option followed by Internet-of-things, Big data, Digital twin, Artificial Intelligence, Machine learning, Cyber-Physical Systems, Blockchain, and Industry 4.0 titleabstract-keyword search option. The time-frame was limited from 2010 to 2021 (current). The authors selected only the articles that were published in English journals and excluded review papers. Conference papers were included because of their novelty, latest findings, and research proposals published in the conference proceedings.

The search code applied was as follows:

(food AND supply AND chain) AND TITLE-ABS-KEY (technological AND advancements) OR TITLE-ABS-KEY (internet AND of AND things) OR TITLE-ABS-KEY (big AND data) OR TITLE-ABS-KEY (digital AND twin) OR TITLE-ABS-KEY (artificial AND intelligence) OR TITLE-ABS-KEY (machine AND learning) OR TITLE-ABS-KEY (cyber AND physical AND systems) OR TITLE-ABS-KEY (block AND chain) OR TITLE-ABS-KEY (industry 4.0)) AND PUBYEAR > 2009 AND (LIMIT-TO (DOCTYPE, "ar") OR LIMIT-TO (DOCTYPE, "cp")) AND (LIMIT-TO (LANGUAGE, "english")) AND (LIMIT-TO (SRC-TYPE, "j") OR LIMIT-TO (SRCTYPE, "p")) Results.

Initially, 156 documents were obtained. Then, the duplicates were removed, and authors thoroughly read the title and abstract of all the papers to scrutinize and bring down the number to 112 final datasets. A detailed methodology with a schematic is shown in Figure 1.

The dataset was snowballed to obtain results such as publication trends and distribution, source of publication and related technological concepts they primarily focused on, research work that was highly cited in this area along with the current number of citations, FSC properties, and percentage of research work plotted against the respective technological tool inclusion, department-wise categorization, and related research work, and barriers in technological adoption in FSC with future research trends. Insights of systematic analysis assisted in systematizing and structuring the dataset and understanding the current trends and challenges in FSC.

A bibliometric analysis on keyword coupling (food safety, quality, and sustainability) was also performed with the same dataset to interpret the relevance and concentration of research work. The authors wanted to understand different clusters of research work on this area. All indexed keyword coupling was run to retrieve the word cloud and gain an overall idea of the research area targeted over the past decade, and the country-wise link strength and citations were retrieved to understand the research work conducted according to the geographical locations. In addition, the relevance between the publication sources was checked to study how the researchers coauthored and cited other research publications in other journal sources. The main agenda of using both forms of analysis is to reap maximum insights on the topic of study, identify research gaps to answer for the research questions, and to propose future research directions.

**Figure 1.** Methodology.

#### **4. Results**

The percentage of type of publications and datasets distributed was computed and projected in Figure 2. Conference papers accounted for 31% of the total publications. Figure 3 portrays the trend in publication. There is gradual rise, generating a good number of publications especially in the years 2017 and 2019.

**Figure 2.** Dataset Distribution.

Table 1 displays the number of publications relating to technological advances in research work over the timeframe. The *International Journal of Production Research* and *Journal of Cleaner Production* have the most publications in the area of AI and blockchain applications in FSC. Furthermore, the top 15 most highly cited research works are tabulated in Table 2. The technological evolution has occurred gradually from applying RFID, IoT, blockchain, and AI.

**Figure 3.** Trend in Publication.



The abstract, title, and full text (only those available) were thoroughly reviewed by the authors to retrieve information on the percentage or volume of technological tools adopted in FSC-based research, which is shown in Figure 4. The inferences show that IoT and big data have been extensively applied; however, AI, ML, cyber-physical systems, digital twin, and blockchain technology still need more attention to discover further implications and benefits for FSC. Later, authors divided FSC based on food quality, safety, and waste and found out the relevant technological adoptions to meet the research objectives which is portrayed in Table 3. The findings reveal that IoT-assisted blockchain technology, RFID integrated with IoT, artificial intelligence, and machine learning were applied to improve food safety and quality.


#### **Table 2.** Author (top fifteen) vs. Problem addressed vs. Citations.

**Figure 4.** Percentage of Technological Tools.



 The FSC was classified into Production and Processing, Food Tracking and Traceability, Warehousing and Packaging, Logistics Branding, Marketing & Sales, and the corresponding technology applied. This classification was performed to obtain an in-depth idea of the technological tools and advancements at different stages of the food chain, starting from raw materials and ending with finished goods. This information is tabulated in Table 4. The results show that more research has been conducted on food traceability and tracking in recent years.

3

3

3

3

3

3

3

3


#### **Table 4.** Department-wise Categorization of Technological Tool Adoption.

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

#### **5. Bibliometric Analysis of Food Safety, Quality, and Sustainability Using Keyword Coupling**

A bibliometric keyword coupling was conducted using the Vosviewer software on the dataset with 978 keywords. The number of keyword repetitions was set at three, in which 83 keywords met the criteria. The keyword nodal burst was separately captured from the bigger image and projected as Figure 5a–c to visualize food quality, safety and waste (sustainability). The authors selected the sustainability keyword-based nodal image in the keyword coupling related to food waste, since the waste node was much smaller and meagerly relevant compared with the other bibliometric, full-factorial coupling clusters.

**Figure 5.** *Cont*.

**Figure 5.** (**a**) Food Quality, (**b**) Food Safety, (**c**) Food Waste (sustainability).

The keyword bibliometric coupling results show the relevance of the research area (keyword) and the intensity of work done based on the distance between them and the size of the node, respectively [43,75]. Inferences can be retrieved by identifying the gaps and future research work. The results from 5a reveal that smart contracts have been deployed in the food industry to maintain food quality. IoT, embedded systems, and Dematel Fuzzy logic algorithms are the primary tools related to food quality maintenance. Whereas IoT, AI, radio frequency identification device (RFID), embedded systems, and big data have all contributed towards maintaining food safety, as shown in Figure 5b. However, IoT and big data are in different clusters with less relevance. This indicates that more research is needed to understand the challenges and drivers of those technological adoptions in FSC from food safety. The same tools are also assisting in sustainability but are seen even farther apart in the nodal burst. The interpretation of the results from Figure 5c keyword burst shows that food waste or sustainable norms of food production and logistics are still in their infancy.

#### *Indexed Keyword Coupling*

Another set of keyword couplings on the indexed keyword set was conducted to visualize the overall keyword cloud. The minimum number of keyword occurrences were three, and 59 met the threshold out of 741 keywords. The indexed keyword coupling based on text mining has been shown in Figure 6. Five clusters have been identified from the word could. Artificial Intelligence, decision support systems, and big data (data mining) are grouped along with the agricultural systems and food traceability. A separate cluster has been generated for food storage and traceability related to food safety. RFID, IoT, blockchain, and agricultural robots are grouped in a separate cluster. Sustainability and strategic decision making for risk assessment seem to be very close and relevant.

The dataset was further reviewed to generate country-wise relevance, number of documents, and total citations per country. A maximum number of research and citations in FSC and technological adoption has been seen in the United Kingdom, followed by India, China, Turkey, and United States. The minimum number of documents and citations per country was fixed as two. A total of 26 countries out of 33 met the criteria as tabulated in Table 5. This

inference is crucial to finding out from which countries researchers and institutions contribute more towards FSC and push other researchers to discover their objectives.

**Figure 6.** Indexed keyword coupling.

**Table 5.** Research Link Strength and Citations between Countries.


Later, the bibliometric coupling on sources was conducted with one article having a minimum of 10 citations. Out of 90 sources, 29 were the most relevant, which is clearly shown in Figure 7. The larger the nodes, the greater the research volume, and the closer the nodes more relevant the research work. *International Journal of Production Research, Journal of Cleaner Production, Sustainability, Industrial Management and Data Systems*, *Information*

*Systems Frontiers*, and *Food Control* are the journals that have been extensively published in these areas.

**Figure 7.** Bibliometric coupling of Journal Sources.

#### **6. Discussion**

Global warming, population growth, industrialization, and the need for sophisticated food systems are all being addressed by innovation. Applying technologies in monitoring ecological effects, smart farming, and value addition for future smart value chains has a tremendous and intriguing perspective. For predicting and forecasting crop cultivation, reaping time, and grade, technologies such as AI are employed to find in-time conveyance and optimized market outreach. This systematic review and bibliometric analysis yielded a set of research implications, which the author discusses in depth in the sections below.

#### *6.1. Effect of Current Pandemic on FSC*

The COVID-19 outbreak gave birth to a new phase in the food sector and supply chain. The repercussions on humankind, the economy, and food safety are still being worked out. Food scientists and experts face numerous issues, including securing food safety, identifying SARS-CoV-2 locations where food is produced, processed, and distributed, and effectively sanitizing surfaces and working areas. More precautions are required as we progress to the final stages of the supply chain, as more people are involved in the process. Food monitoring and surveillance would become increasingly reliant on the development of effective bioanalytical technologies [76].

The pandemic is responsible for rapid shifts in the foodservice to retail food patterns requiring flexible FSC. Potential long-term changes in the supply chain include greater food supply-chain automation and digitization. In addition, technological investments in online delivery infrastructure have changed retail food landscapes. Nonetheless, the danger of labor scarcity due to worker sickness, self-isolation, or movement constraints has critical consequences and makes FSC more vulnerable. Significantly, in the meat processing and general food packaging industries, the demand has increased substantially [77].

The working atmosphere experiences a complete transformation where most of the work is from home, depending on digital communication and contactless electronic communications. Therefore, the technological inclusions in the food system that have been incorporated, especially in areas such as quality control, verification, and certification, have improved FSC. However, the physical inspection of food items during the packaging and logistics procedures are still facing challenges due to disruptions in the supply-chain footprints [13,78].

Policy guidelines and operations are being amended continuously. There is a greater need to tap and leverage the full capability of IR 4.0 technological tools and protocols to overcome the challenges due to pandemic disruption. Truck routes can be optimized, warehouse locations can be divided and scattered, we could rely on locally grown crops, implement agile and lean methods in agriculture, and most importantly the supply chain footprints should be planned to create supply-chain viability.

#### *6.2. Technology and Food Sustainability*

The current scenario necessitates the convergence of appropriate supply-chain systems with industry 4.0 to maintain sustainability. An intelligent food-production system can effectively address challenges in food safety, security, control, and perishability [17]. One of the biggest reasons for the world's existing sustainability challenges might be attributed to the lack of potential to incorporate technological advancements effectively [79].

Given the perishability of food and the importance of food safety in agricultural goods, a better technology-driven strategy is required at every stage of the food supply chain during processing and manufacturing to avoid waste and assure high-quality end products [17,80]. To bolster these facts, Belaud et al. (2019), [81] developed a big-data integrated food supply-chain design for the bioconversion of lignocellulosic biomass, creating environmental sustainability in the agricultural waste valorization domain. These technologies directly and favorably impact traceability, compliance, and coordination between FSC actors and their adoption-intention decision processes that generate scalable, interoperable, and cost-effective architecture for supply-chain integration and sustainability [82].

#### *6.3. Scope for Circularity in Food Supply Chain and Waste Management*

Many research projects are focused on reducing food waste. Product deterioration and decomposition were identified as three main sources of food waste during logistics [83–85]. Food organizations are trying to adopt circular economy strategies to improve supplychain ecological stability. However, from the perspective of underdeveloped nations, the adoption of circular economy and sustainability elements is more complicated than in rich countries. An excellent sustainable strategy shall rewrite poor government policies, lack of technology and practices, and lack of awareness and education. These are among the main obstacles to a successful circular economy-led sustainable supply-chain integration [86].

Green and sustainable supply-chain management methods have emerged in recent decades to incorporate environmental concerns within organizations by avoiding unexpected negative environmental repercussions due to consumption. Parallel to this, the circular economy concept has gained traction in the literature and in practice in industrial ecology. The circular economy pushes the bounds of environmental sustainability by emphasizing the idea of designing the products so that there are viable linkages between ecological systems and product consumption [87].

#### *6.4. Technological Adoption in FSC and Challenges*

Effective management of food safety and security, demand and supply shortages, quality of products, and traceability, can bring economic and social progress in the food sector. Technological tools provide viable and protracted platforms to reduce human intervention and error [88]. Reconceptualizing supply-chain design and operations with

the help of digital technologies helps in overcoming the barriers in FSC [89]. However, very little research has been conducted on the factors that affect these technologies' adoption to attain supply chain 4.0. More research into the perceived drivers and hurdles to implementing supply chain 4.0 in the context of FSC is required. The significant challenges and barriers are supply–demand imbalance, rapidly changing customer expectations, legal ramifications, cost optimization, and lack of organizational collaboration [90].

The introduction of blockchain technology resolves many challenges related to food integrity, traceability, and audit [80]. Casino et al. (2021) [4] stated that upstream and downstream supply-chain players are pushed to store and manage traceability-related data to provide proof of regulatory compliance to government authorities. Tian et al. (2017) [50] developed a food supply-chain traceability system for real-time food tracing based on HACCP (Hazard Analysis and Critical Control Points), backed by blockchain and the Internet of Things, which provided an open, transparent, neutral, reliable, and secure information platform for all supply-chain members in FSC. Chen et al. (2017) [91] introduced a unique, intelligent, predictive food traceability with a cyber-physical system coupled with simulation modelling by combining intuitionistic-based fuzzy case-based reasoning with enterprise architecture and value stream mapping. The CPS-based food traceability system was utilized to identify traceable objects that are reactive to a broader range of intelligent food traceability using a novel approach for traceability performanceprediction behavior.

IoT can give concrete and commercial benefits to FSC, hence improving the efficiency and productivity of operational procedures. However, it is increasingly difficult for retailers to adapt their marketing strategies to shifting consumer behavior as the food retailing industry becomes more complicated and flexible. Internet of Things (IoT) is intended to assist businesses in checking the quality of food products, planning waste management for things beyond their shelf life, managing shop temperatures and other equipment that reduces energy use. As a result, the adoption of IoT is currently in infancy, despite its enormous potential [57]. Cyber-physical systems (CPS) have now been introduced to take care of food traceability from a future internet perspective to display intelligent behavior such as smart predictive business practices in the FSC. Nonetheless, the CPS-based food traceability system faces several new issues, including communication efficiency, heavy capital investment, and system architecture requirements [91].

#### *6.5. Role of Technology in Food Relationship Strategies*

Horizontal collaboration and relationship policies between FSC players are the need of the hour, where there are very minimal supply chain footprints and routes, especially during this COVID-19 pandemic. Therefore, proper collaboration and cooperation strategies in food supply chains can improve resource usage and market governance. Furthermore, they can assist in enhancing the FSC resilience and all three different dimensions of sustainability [92,93]. Effective relationship strategies through horizontal and vertical collaborations improve cost and quality in FSC [94,95].

Designing processes to jointly reap the benefits via developing goals and also investing in capabilities and assets are very essential. Technological implementation will ease the planning and goal-sharing setup in FSC. State-conflicting goals should be avoided by framing better relationship strategies. Blockchain-based smart contracts in the food supply chain and IoT-assisted big-data cloud technology can help overcome this challenge by setting up secure contracts between stakeholders and increasing FSC integrity [96].

The blockchain smart contract would have an RFID identifier preinstalled that would retrieve information on the area, state, nation, time related to product packaging, storing, transportation, and product quality. An ID tag is a setup in the RFID label that would be integrated with the blockchain to store permanently immutable information for secured time-stamped transactions. Collaboration and establishing business contracts among the food supply-chain players to incorporate food relationship strategies is eased by this protocol [97]. Furthermore, technological platforms can be shared between competitors to enable an effective downstream horizontal collaboration through mutual trust and benefit sharing [98].

#### *6.6. Food Supply Transformations through Technology*

Achieving food-system sustainability is a global concern, especially knowing how in-parallel food supply transformation could be accomplished. The practically feasible role of technology and human engagement with agricultural systems are pondered to streamline this food supply-chain transformation. Food sustainability, integrity, traceability, safety, waste management, and pandemic disruptions are major elements in the FSC to be considered for transformation and more resilience [12,99,100].

Technology adoption in FSC creates transformation both in the quality and safety of food products. Moreover, technology has been adopted to improve resource efficiency and productivity in food systems. This has reduced agricultural raw material inputs to reduce environmental externalities. Many farms across the world are applying big data and data analytics in equipment maintenance, field mapping, and other operational activities to optimize irrigation to improve the productivity of agricultural practices. Additionally, digital-twin technology-based geographical information systems (GIS) are adopted to perform precision agriculture that allows the utilization of sensors to optimize the use of pesticides, fertilizers, and water. Moreover, other decision support systems help farmers to maximize production efficiency while minimizing production costs and the environmental footprint of their operations. These aspects serve as a building block for the transformation of food systems [101].

#### **7. Future Research on Technological Inclusions for Food Supply-Chain Transformation and Innovation**

After a systematic literature review and bibliometric analysis, authors have accumulated insights on the future research scope and direction. More research should be focused on innovating agricultural farming, production, and processing with the help of smart supply chains and digital technologies. There are significant research opportunities if artificial intelligence and machine learning are applied to control food transport optimization issues, demand-forecasting, prescriptive shipping technologies for perishable food products, and organizing safety and quality in the food chain. The percentage of customer satisfaction should be kept as a key performance index during the integration of technological tools and FSC. Blockchain-based smart contracts can be built to complete state-of-the-art functional and purpose-driven supply-chain and financial transactions. Moreover, the food supply chain needs to be strengthened more from all three facets (food quality, safety and sustainability) in order to fight the COVID-19 pandemic disruptions. Additionally, IoT-assisted big data can build horizontal collaborations that improve food relationship strategies.

Government policies, approvals, and audits can be digitalized using the blockchain and IoT to increase FSC resilience. Blockchain platforms can also create traceability certificates capturing all the supply chain footprints. Cyber-physical systems can directly help in food processing and packaging in this and next decade, where fewer human interactions are desired due to the pandemic. The quality of the FSC from a micrologistics perspective can be improved using cyber-physical systems and smart robotics in the food processing and packaging area. Blockchain and big-data-driven technology can assist farmers in practicing responsible procurement to maintain sustainability standards, both environmentally and economically. A complete food supply transformation-based operational paradigm is shown in Figure 8. After a detailed review of the dataset, the authors propose related technological interventions that are required at different stages of the FSC. It displays barriers and challenges at the different echelons of FSC and the technology tools that can be applied to overcome them and create scalability for more supply chain 4.0 drivers in FSC.

**Figure 8.** Food Supply Chain 4.0 Operational Paradigm.

The costs associated with FSC such as logistics, freight, energy, fuel, workforce, and capital investment in technology should be kept to a minimum to suppress the bullwhip effect in the chain. IoT-assisted big data can help in this aspect by creating cost patterns in the data warehouse and showing the predictive and prescriptive solutions for better decision making using machine-learning algorithms. In addition, a high level of quality and safety is needed for final food products at all times, both globally and locally. Enhancing the visibility and interaction in the FSC, a business can witness significant gains.

#### **8. Conclusions**

This study aimed to systematize the previous literature on FSC and the application of IR. 4.0 tools, and review how the past research has been focused on counteracting the disruptions in FSC. More problems need to be addressed regarding how to effectively integrate one or more tools to reap maximum benefits. Very few studies have applied blockchain (integrity, security), artificial intelligence and machine learning (error-free prescriptive platform), digital twin, or cyber-physical systems within the scope of the study. Additionally, there is a need to build more digital support systems for FSC to improve decision making, especially within pandemic conditions.

More studies must be focused on avoiding food wastage. However, technical failures in the supply chain eventually result in food waste. The cost of monitoring suppliers makes it difficult for retailers to embrace new and innovative suppliers. More modern automation in food systems has piqued the interest of food manufacturers regarding long-term investment. Unquestionably, the impending food catastrophe cannot be cleared overnight. The apparent benefit of digitization is that it helps to reduce waste that could otherwise be avoided. When one out of every three freight journeys is for food, generating better real-time data to enhance routes and distribution planning is critical. Furthermore, by utilizing digital and automation technologies, food loss may be avoided and costs can be drastically reduced. When real-time data is used with a variety of sustainable indicators, businesses may drastically cut yearly energy utilization.

Future research should be aimed at improving the level of digitalization, marching towards strong traceability systems that can control food advocacy, source, and safety during this pandemic, where counterfeiting and adulteration can more common than usual. Moreover, digitalization offers a complete audit trail of trustworthy information that enables the supplier to enter the supply chain with the capacity to validate the quality of the production and the procedures at all stages, from farm to retailer. More research should be focused on traditional food procurement methods that have spawned both consumer expectations and misconceptions. Consumers should be more informed and educated about food quality and its health consequences. The use of technological instruments reduces waste in FSC, strengthens its resilience, and increases viability. The changing end-to-end business model relies mainly on revolutionary innovation in the food sector. Food safety and advocacy will improve as a result of embracing digitalization, allowing the market to democratize accessibility and experiment. All of this is possible due to the industry's automation, increased efficiency, improved consumer knowledge, and support for important food production and consumption changes.

Furthermore, achieving transformation in the food system would need a significant shift in attitudes, as well as the roles and duties of public sector actors versus corporations in determining food demand. This can be achieved by properly planning horizontal collaboration protocols in FSC. Economic development, human health, and planetary health are all dependent on food systems, and getting all three right is critical. They are intertwined and have a significant impact on one another. Every nation must conceive prospective future possibilities in which everyone consumes adequately, based on food systems that are ecologically, economically, and socially viable. Local and national perspectives on how such food systems would appear in their higher prevalence should guide policy goals intended to achieve long-term transformation.

**Author Contributions:** Conceptualization, methodology, validation, J.P.; software, data curation, writing—original draft preparation, A.Z.A.; supervision V.P.K.S.; revision and supervision, A.K.O.; reviewing and editing, review protocol, investigations, S.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** Authors would like to thank the reviewers for their constructive comments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Logistics* Editorial Office E-mail: logistics@mdpi.com www.mdpi.com/journal/logistics

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com