It is often assumed that a fourth agricultural revolution is starting, in which digital technologies will play a critical role in transforming agricultural production – comprising crops, livestock, aquaculture and forestry – in a move towards increased efficiency and sustainability. These technologies include artificial intelligence (AI), drones, robotics, sensors and global navigation satellite systems (GNSS), as well as other digital tools that help automate diagnosis, decision-making and performing in various agricultural activities, allowing increased precision and efficiency.2 Some of these technologies are commercially available, while others are approaching readiness levels.26
Various scenarios for agriculture in the coming years and decades point to a likely rise in the use of different digital and automation technologies.27, 28 In recent years, the vast proliferation of hand-held devices (e.g. mobile phones and smartphones, sensors, internet of things [IoT] devices) is clearly visible, and is largely the result of improved access to mobile networks and expanding internet coverage, even in the world’s remotest regions. For example, in 2020, 69 percent of the population in Latin America and the Caribbean, 64 percent in Pacific Asia, and 45 percent in sub-Saharan Africa had acquired a smartphone, and these figures are expected to increase to 81 percent, 79 percent and 67 percent, respectively, by 2025.29 This is the result of massive investments in infrastructure by both governments and the private sector. For example, Google is investing in Africa’s first subsea internet cable through its Equiano programme.30
There follows a presentation and analysis of the potential of these technologies to transform the landscape of motorized mechanization and agricultural operations in general.
Digital technologies are transforming conventional agricultural machinery
Policymakers and international organizations increasingly view digitalization as a game changer in the agriculture sector. Central to most digital technologies is the possibility to collect and exchange data to support decision-making by agricultural producers or other stakeholders and, ultimately, to enhance effectiveness and efficiency.31, 32 In recent years, these digital technologies and services have received significant attention from donors, research centres and development agencies.29, 33, 34, 35 They are increasingly incorporated in motorized machinery, potentially transforming its use: agricultural operations are performed with more efficiency and precision, and access to agricultural machinery is extended to new regions or socioeconomic groups, such as small-scale producers.
Many of these technologies are based on applications operated by a smartphone, or via a call or messaging service. Shared asset services are a subcategory of digital services, with significant potential to expand access to motorized mechanization, connecting owners of equipment (e.g. tractors or drones), and sometimes also operators, with agricultural producers who need such equipment. Agricultural producers pay the owner per hour or per area serviced, and a percentage or fixed fee goes to the matchmaker. The best known example of a shared asset service is Hello Tractor (operating in seven African countries as well as in Bangladesh, India and Pakistan).1 See Box 3 for two successful cases in some African countries and Myanmar.
Box 3Digital tools for improved access to mechanization services
Digital tools based on the Uber taxi model are on the rise and promise to reduce transaction costs for tractor services. TROTRO Tractor in Ghana, and Tun Yat in Myanmar hire out machinery and share services through a digital platform and mobile phone services. These tools demonstrate real potential for inclusive agricultural mechanization.
TROTRO Tractor matches small-scale producers with the agricultural machinery they require, primarily tractors, and with the owners of that machinery, through a digital platform accessed via smartphone apps, as well as through unstructured supplementary service data (USSD) for users who do not own a smartphone. Currently TROTRO Tractor has 75 000 farmers registered across Benin, Ghana, Nigeria, Togo, Zambia and Zimbabwe. It relies on both business-to-client and business-to-business relationships, retaining a commission percentage on the cost of the service.
In addition to tractors (offering any type of service from ploughing to harrowing, and from planting and seeding to spraying) and combine harvesters, the TROTRO Tractor platform also connects producers with drone owners who offer their services for mapping and herbicide spraying. There is growing demand for drone-based mapping as farmers appreciate the importance of land tenure and understand that accurate land data may be crucial when requesting financial services from banks or insurers.
Tun Yat provides similar tractor services through a smartphone app, specifically targeting small- and medium-scale farmers, with a focus on women (representing 30 percent of clients) and youth (with 25–30 percent of clients under the age of 30). Tun Yat owns five tractors and five combine harvesters and offers a range of mechanization and matchmaking services to more than 20 000 customers. Services include ploughing, land preparation, seeding, combine harvesting with different headers for different types of harvest (e.g. mung beans or maize), and picking (e.g. sesame or groundnut). Most customers are small-scale producers with landholdings under 2 ha, who are especially in need of reliable and affordable mechanization services.
The Tun Yat business model embraces diversification, with services including resale of inputs (e.g. fertilizer), credit brokerage, and laser levelling to assist farmers in flood-prone areas who need to level farm plots and develop drainage. It also offers direct purchase from farmer groups of raw material, which is then processed into snacks and sold at convenience stores.
In synthesis, the Uber-like business model is advantageous both for farmers who do not own tractors and for equipment owners; the latter can maximize, closely monitor and plan machinery use and fuel consumption, offering competitive rates to a broader customer base.
SOURCE: Ceccarelli et al., 2022.2
The main benefit of these shared asset services is an improved cost–benefit ratio: farmers gain access to the equipment they need without having to buy it, while the fees paid make the equipment more cost-effective for the owner. These asset services are especially important in sub-Saharan Africa, where ownership is extremely limited (see Box 2).
Another group of digital services are equipment-monitoring solutions, that is, simple applications that automate the operation of equipment such as irrigation pumps,36, 37 or GNSS devices to track movements of, for example, equipment or animals. These types of services are seen as the first smart farming solutions to emerge for low- and middle-income countries.38 More advanced services include the IoT solutions used, for example, to monitor and sometimes (partially) automate decisions concerning the care of crops, livestock or fish in order to improve diagnosis, decision-making and performing. This in turn leads to enhanced precision, improved efficiency and increased productivity, while reducing drudgery. A concrete example of IoT use for precision agriculture comes from China, where it supports an integrated system of automatic remote sensing, early warning and microspray irrigation for tea production; changes in environmental conditions are detected, timely warnings are provided, and irrigation is triggered automatically, as and when necessary, thus avoiding damage from heat, cold or drought.39
The transformation by digital technologies of the use of motorized machinery such as tractors and harvesting equipment is somewhat limited, especially in low- and lower-middle-income countries.1, 2 On the other hand, the organizational models for the use of motorized machinery are undergoing significant changes. There is an increasing focus on shared rather than individual ownership of machines by producers in low- and middle-income countries. Asset sharing has existed for a long time, but with limited success due to, for example, distrust between farmers, operators and machine owners, and issues related to machine maintenance. More recently, IoT and GNSS solutions, although still very limited among small-scale producers, are being widely adopted by service providers (including those mentioned in Box 3). By facilitating monitoring of the machinery, they enhance transparency and trust between service providers and users. Perhaps the most important change is the embodiment of traditional mechanization equipment with IoT devices (e.g. a combination of motorized harvesting equipment from a hiring service, GNSS data and a trained operator to drive a tractor), which can result in more effective use of machines, as well as higher yields.1
The potential of digital technologies for non-mechanized precision agriculture
The previous section described how digital technologies can transform the landscape of agricultural machinery, making mechanization both more precise and more accessible. Nevertheless, the adoption of motorized agricultural mechanization is still limited in many low- and middle-income countries, especially in sub-Saharan Africa. There is growing research on precision agriculture for non-mechanized production and its adoption is increasing.40, 41, 42 Methodologies for manual site-specific fertilizer application were developed a long time ago – for example, variable rate technology (VRT) for fertilizer on rice43 – while the AgroCares hand-held soil scanner is available in several low-income countries in Africa and Asia.44 Non-mechanized farms in Africa and Asia are adopting uncrewed aerial vehicle (UAV) services (also known as drones), while GNSS can be used on non-mechanized farms to map field boundaries and establish land tenure.45
However, there is a lack of information on adoption levels; it is not clear how many agricultural producers actually use digital technologies.46 Results from two technical studies – commissioned for this report1, 2 – indicate that at the field level, a variety of digital tools and remote sensing and mapping technologies are increasingly used by small-scale agricultural producers and pastoralists across the world (see Box 4). Smartphones, with diverse sensors and high-resolution cameras built in, are the most accessible hardware for everyone in low- and middle-income countries today. In combination with apps embedded in smartphones and suitable interfaces, they can already make available highly useful innovations appropriate to the context of low- and middle-income countries and small-scale agriculture and have the potential to make a real difference. One example is GoMicro: through a microscopic lens clipped onto a phone camera, combined with AI, it supports the rapid diagnostics of pests and diseases,47 and assists efficient and accurate quality control and grading of agricultural products such as cereals and grains, fish, fruits and vegetables.1 There are other digital solutions involving satellite or drone data (e.g. on yields, soil conditions and plant health) analysed by an algorithm; the results can be used to validate data shared by agricultural producers (based on observations and experience) or to provide advice to producers.1
Box 4Digital tools not linked to mechanization – disembodied solutions
Disembodied digital solutions (see Glossary) are not linked to mechanization. They are primarily software-based solutions that do not rely on the use of agricultural machinery. Instead, they require limited hardware resources, generally in the form of a smartphone, tablet, software tool (e.g. advisory apps), farm management software or online platform. This sets them apart from embodied digital solutions, where digital tools are combined with machinery to interact with the environment.
Disembodied solutions may include remote sensing, but limited to data for decision support and scouting. These are increasingly used across the globe as illustrated below by examples from across the world. The South African company, Aerobotics, operates in 18 countries, offering disembodied solutions with an uncrewed aerial system (UAS) and remote sensing for decision support to growers of fruits and nuts. The technologies allow early detection of pests and diseases, enable timely monitoring of water, fertilizer and nutrient requirements, and facilitate yield management.
In Morocco, SOWIT offers disembodied solutions using remote sensing, UAS for imagery collection, and machine learning based on data from field or weather databases. The technologies can be applied to fruit trees, cereals and rapeseed and they inform farmers of irrigation and fertilization requirements, estimate yield, monitor the dry matter content of forage, and carry out plot inspections.
In Nepal, Seed Innovations offers an Android application for farmers to use satellite-based analytics, global navigation satellite systems (GNSS) and artificial intelligence to monitor crop performance – including identification of water and nutrient deficiencies or surplus, and of pest and disease threats – and access and exchange agronomic information.
Based in Fiji, TraSeable Solutions has 2 000 active customers across seven Small Island Developing States in the Pacific. The company offers two main solutions. The first is a mobile app that informs farmers about the agriculture sector, records and manages farm data, and keeps track of resources, inventory, sales and expenses. The app also helps create market linkages between agriculture value chain stakeholders. The second focuses on fisheries, specifically tuna. It involves the tagging and tracking of individual tuna along the value chain from landing through to distribution. This solution also helps to manage fleets by providing information on the crew, and on operation and maintenance costs. In addition, it supplies tuna harvest details, including trip information, catch log sheets, fishing ground analytics and reporting services.
In Peru, Coopecan offers digital services along the whole Alpaca fibre value chain. A range of technologies provide digital solutions for, among others, pasture management (satellite imagery), animal health (animal tags), and fibre processing and export sales (blockchain technology). In addition, technical assistance is available for breeders needing support in herd management (e.g. regarding animal health status) or in management of natural pastures (increasingly degraded due to excessive grazing). These services are complemented by capacity building on how to use the solutions, and a traceability system that certifies production in terms of animal wellness, fibre quality and environmental and social responsibility, leading to better working conditions, fair pay and improved animal welfare.
Finally, Agrinapsis, operating in Bolivia (Plurinational State of), Costa Rica, Ecuador, Guatemala and Mexico, is a social media platform specializing in agriculture. Managed by the Inter-American Institute for Cooperation on Agriculture, it facilitates the exchange of common knowledge among small-scale producers. The crowd-sourced information is verified and rated by all its customers and, if flagged as doubtful or of poor quality, a technical team checks and improves it. Agrinapsis enables e-commerce targeted at small-scale producers, who can sell their produce or buy inputs that respond to environmental concerns.
Digital solutions deserve the attention of policymakers and international organizations. More research is needed to tailor them to the needs of small-scale producers in low- and middle-income countries, especially in least mechanized ones, such as in sub-Saharan Africa.1 Research and experience indicate that they enable site-specific crop management, which in turn can improve yields and reduce inputs on non-mechanized farms. However, two constraints are worth mentioning. First, digital technologies may be too expensive for small-scale producers, given the current costs of the equipment. For example, hand-held nitrogen sensors are priced between USD 300 and USD 600, which is excessive for a small farmer who only wishes to use the sensor a few times a year,48 while the more sophisticated AgroCares scanner, which provides information on a wider range of soil nutrients, sells for over USD 3 000. Second, producers need to learn how to use the technologies; without the know-how, incorrect implementation can lead to undesired results, such as increased use of inputs.
There are still places, primarily in sub-Saharan Africa, where smartphones are outside the reach of small-scale producers and rural populations.1, 2 Data from 2020 also reveal a substantial rural–urban divide in terms of internet access in developing countries: 65 percent of the population in urban areas have access versus only 28 percent in rural areas.49 The evidence points to high cost as a fundamental factor hindering the adoption of these technologies by small-scale producers, despite the significant potential for improving productivity. This suggests that donor-subsidized, low-cost access to digital technologies by small-scale producers may not be viable.
Therefore, more – and more diverse – efforts are needed to make these tools more accessible. Shared asset service providers, mentioned above, are one solution and they already offer a range of machinery appropriate for both small- and large-scale farms. However, the low level of digital literacy among agricultural producers may also be a significant factor in the slow adoption of digital tools. For this reason, SMS text messages, interactive voice response (IVR) and unstructured supplementary service data (USSD) services are used to communicate with small-scale producers in many African countries. For example, ICT4BXW and Justdiggit – both operating in sub-Saharan Africa – initially adopted advanced technologies such as smartphones, but then decided to use simpler means (SMS, USSD and IVR) due to low smartphone penetration and the low level of digital literacy in the region.1
For as long as farmers have poor digital literacy, intensive and continuous technical support and information about digital technologies must be provided. Although basic mobile phones are now accessible to almost everyone, smartphones remain limited in sub-Saharan Africa. Therefore, phones need to be used in combination with advisory services based on satellite intelligence tailored to producers’ needs (e.g. on water sources and grazing grounds for pastoralist livestock keepers, and on disease outbreaks for banana farmers).1
