Search

Plant diseases affect the quality and quantity of agricultural products and have an impact on food safety. These effects result in a loss of income in the production sectors which are particularly critical for developing countries. Visual inspection by subject matter experts is time-consuming, expensive and not scalable for large farms. As such, the automation of plant disease detection is a feasible solution to prevent losses in yield. Nowadays, one of the most popular approaches for this automation is to use drones. Though there are several articles published on the use of drones for plant disease detection, a systematic overview of these studies is lacking. To address this problem, a systematic literature review (SLR) on the use of drones for plant disease detection was undertaken and 38 primary studies were selected to answer research questions related to disease types, drone categories, stakeholders, machine learning tasks, data, techniques to support decision-making, agricultural product types and challenges. It was shown that the most common disease is blight; fungus is the most important pathogen and grape and watermelon are the most studied crops. The most used drone type is the quadcopter and the most applied machine learning task is classification. Color-infrared (CIR) images are the most preferred data used and field images are the main focus. The machine learning algorithm applied most is convolutional neural network (CNN). In addition, the challenges to pave the way for further research were provided.

This paper presents a conceptual framework that reflects the current state of thinking on tokenizing circularity in agri-food systems. The framework is built upon classifications of tokens and the key principles of circular economy and shows how tokenization can support circularity in agri-food systems through the flows of information and flows of value. Based on an integrative review of literature on tokenization, blockchain and the circular economy and multiple case studies in the agri-food domain, we show the relevance of tokenization to the circular economy in three ways: 1) enhancing traceability of physical and digital objects in supply chains; 2) improving transparency and credibility of circularity claims; 3) facilitating collaborative business ecosystems with incentives for more circular production and distribution. Based on the framework, we derive important research questions for future research agenda on tokenizing circularity in agri-food systems.

PurposeTraceability of food is of paramount importance to the increasingly sustainability-conscious consumers. Several tracking and tracing systems have been developed in the AgriFood sector in order to prove to the consumers the origins and processing of food products. Critical challenges in realizing food's traceability include cooperating with multiple actors on common data sharing standards and data models.Design/methodology/approachThis research applies a design science approach to showcase traceability that includes preharvest activities and conditions in a case study. The authors demonstrate how existing data sharing standards can be applied in combination with new data models suitable for capturing transparency information about plant production.FindingsTogether with existing studies on farm-to-fork transparency, our results demonstrate how to realize transparency from field to fork and enable producers to show a complete bill of sustainability.Originality/valueThe existing standards and data models address transparency challenges in AgriFood chains from the moment of harvest up to retail (farm-to-fork) relatively well, but not what happens before harvest. In order to address sustainability concerns, there is a need to collect data about production activities related to product quality and sustainability before harvesting and share it downstream the supply chain. The ability to gather data on sustainability practices such as reducing pesticide, herbicide, fertilizer and water use are crucial requirements for producers to market their produce as quality and sustainable products.

One of the key elements of precision agriculture is the farm management information system (FMIS) that is responsible for data management, analytics and subsequent decision support. Various FMISs have been developed to support the management of farm businesses. A key artefact in the development of FMISs is the software architecture that defines the gross level structure of the system. The software architecture is important for understanding the system, analysing the design decisions and guiding the further development of the system based on the architecture. To assist in the design of the FMIS architecture, several reference architectures have been provided in the literature. Unfortunately, in practice, it is less trivial to derive the application architecture from these reference architectures. Two underlying reasons for this were identified. First of all, it appears that the proposed reference architectures do not specifically focus on FMIS but have a rather broad scope of the agricultural domain in general. Secondly, the proposed reference architectures do not seem to have followed the proper architecture documentation guidelines as defined in the software architecture community, lack precision, and thus impeding the design of the required application architectures. Presented in this article is a novel reference architecture that is dedicated to the specific FMIS domain, and which is documented using the software architecture documentation guidelines. In addition, the systematic approach for deriving application architectures from the proposed reference architecture is provided. To illustrate the approach, the results of multi-case study research are shown in which the presented reference architecture is used for deriving different FMIS application architectures.

Nowadays, we produce and consume a large volume of products giving insufficient regard to what happens to the products before, during and after use or consumption. Globally roughly one-third of food produced is lost or wasted each year. As the world population and per-capita income increase, the problem gets worse. This lifestyle based on “linear” economy has become untenable, and therefore the adoption of a system of business processes that implement circular economy principles approach is proposed. However, it is not clear to what extent the current practice of documenting business processes in supply chains, the SCOR (Supply Chain Operations Reference) model, reflects and captures the circular economy practices. In this study, we reviewed the state-of-the-art literature to identify to what extent the SCOR model has been used in connection to circular economy and to what extent SCOR supports “circular” business processes.

Nowadays, different types of farm management information systems (FMIS) are being used in practice in several sectors of farming, such as dairy, arable, fruits, vegetables, and meat farming. The goal of this research is to identify, evaluate, and synthesize existing FMISs in the Dutch dairy sector and present the state-of-the-art. We performed a multivocal literature review (MLR) to find sources both in scientific and grey literature. A grey literature search was adopted because most of the FMISs were not reported in the scientific literature. To support and improve the effectiveness of the MLR process, an online survey was first sent to Dutch dairy farmers to identify the FMISs that are being used in practice. With the help of the MLR process, we identified 50 FMISs used by Dutch dairy farmers. We identified 33 features of these FMISs and listed the advantages and disadvantages of the FMISs. Copyright: 2022 by the authors. Licensee MDPI, Basel, Switzerland. Funding: Authors thanks to the Molde University College-Specialized Univ. in Logistics, Norway for the support of Open access fund. Scopus 2-s2.0-85122665807

Applications of data analytics, and recently machine learning, in pig farming have been investigated in literature and the results indicate great potential for data-driven decision support at various scales of the sector—from farm to the management of entire supply chains. However, there is insufficient overview of the studies conducted so far. Particularly, there is little insight into the extent of studies conducted in the context of actual business cases. In this study we conducted a systematic literature review to shed light on the state-of-the-art knowledge about data-driven decision making in the pig sector. In order to cover both classical data analysis techniques and machine learning, we used two separate search strings to search the literature. The results show that the various attributes of live pigs and slaughter data are used in analytics. Most studies focus on the occurrence and prevention of diseases, followed by DNA-related analysis and the effect of feeding strategies on growth. Among the studies we analysed, there was a large variation in herd size under study. Most studies used a selected group of pigs in an experimental environment; fewer studies used a larger number of pigs. Notably, all studies except two focussed on real-life business contexts where real-time data is used. The application of machine learning, mainly the use of random forest and neural network algorithms, took off since 2018. Current studies focus on isolated and one-off problems, and we suggest future research to consider the complexity encountered in real-life business circumstances and routine decision making through the integration of data analytics within farm information management systems.

Despite increasing global environmental concerns, we continue to consume large amounts of products with little regard to what happens before, during, and after their use. Roughly one-third of the food produced is wasted. Because the world’s population is expected to grow to 10 billion by 2050, adopting circular economy practices will become essential. The transition towards a circular economy requires adopting business processes that support circular economy practices across supply chains. Currently, the SCOR (Supply Chain Operations Reference) model is the most widely used, and widely known, approach for studying and evaluating supply chain business processes. It is, however, unclear to what extent circular principles are included in the SCOR model. Past studies indicating missing processes for circular supply chain management in the SCOR model have made limited efforts in capturing the current state-of-the-practice. We conducted an online survey of 60 companies engaged in 14 different types of economic activities to study the SCOR level 2 business processes adopted in practice. In addition to the 22 level 2 business processes documented in SCOR, we identified six additional level 2 circular business processes that the respondents recognized as being commonly applied within their businesses. The results clearly show that the current SCOR model does not fully represent circular business processes in the state-of-the-practice.

This paper proposes a drone policy framework for developing drone regulations for Africa, expediting the individual countries’ developing drone regulations. This research was conducted following concerns raised by individuals and businesses whose imported drones were seized by customs or had to go through a cumbersome process for approval to use drones for lack of regulations. To derive the policy framework, we matched the provisions of 15 existing drone regulations from Africa to the International Civil Aviation Organization (ICAO) Manual on Remotely Piloted Aircraft Systems (RPAS) using a decomposition modelling technique. The proposed framework is presented as a formal business process model and is evaluated in a multi-case study approach in two different case studies. The results indicate that two key aspects/elements are crucial in the development of a drone framework and thus are included in our framework: (i) stakeholders’ inclusion in the policy development and (ii) regulators liaising with security agencies like the police and immigration officials to implement and enforce developed regulations. Applying the proposed framework in describing existing regulations reveals that 40-85% of provisions in existing regulations conform to the provisions in ICAO’s RPAS manual.