Emerging Tech in Agri - PWC
Emerging Tech in Agri - PWC
Contents
Foreword 3
Executive summary 4
Introduction 5
1.1
Agritech: The situation today 9
2.1
Solving the unit economics of tech adoption 17
for smallholders
2.2
Gender-inclusive digital agriculture: 18
Acknowledging the critical role of women farmers
2.3
Harnessing the power of public–private collaboration 19
2.4
Linking digital and physical channels 20
Contributors 22
Endnotes 23
Disclaimer
This document is published by the
World Economic Forum as a contribution
to a project, insight area or interaction.
The findings, interpretations and
conclusions expressed herein are a result
of a collaborative process facilitated and
endorsed by the World Economic Forum
but whose results do not necessarily
represent the views of the World Economic
Forum, nor the entirety of its Members,
Partners or other stakeholders.
Foreword
Sebastian Buckup
Head of Network and
Sanjeev Krishan
Partnerships; Member of the
Chairperson, PwC in India
Executive Committee, World
Economic Forum
As the world grapples with the challenges of behind. Excluding them would have a directly
climate change, shifts in geopolitical dynamics, negative impact on global food production
pressure on natural resources and rising pollution, levels, especially in developing nations. It is also
ensuring food security for the ever-increasing global imperative to ensure that agritech services are
population becomes more and more important. affordable and easily accessible through both
Transforming agricultural production, using digital and “phygital” (combining human and digital
technology to enhance efficiency and promote channels) means to ensure effective outreach
sustainability, is an essential part of the solution to and adoption of tech at the last mile. Lastly,
this problem. collaboration is the key to unlocking the other
three principles. Given the complexities of the
While agritech technologies have existed for a agriculture sector, working in isolation will not help
decade, they have been seen as point solutions the world community address the global challenge
mainly focused on farm management rather than of food security.
as a package of resources that can address
critical interrelated inefficiencies in agriculture In 2021, the World Economic Forum’s Artificial
value chains from crop planning to consumption. Intelligence for Agriculture Innovation (AI4AI)
It is imperative that agriculture is viewed from a initiative published a community paper
systems perspective, as an interrelated chain of documenting a range of agritech services from
activities and actors. The use of artificial intelligence crop planning to harvesting, and shared a roadmap
(AI), augmented reality (AR), internet of things of how to work collaboratively with stakeholders
(IoT) devices, robotics, blockchain and drones in scaling these agritech services in Indian
is providing tools for farmers, traders, logistics agriculture. We are pleased to present this report
providers and food-processing plants to bring as a sequel to the 2021 community paper. It aims
about this change in the agricultural sector and to provide a macro-level view of the integration of
make it a more agile, better-informed and well- technologies needed to deliver systemic change in
connected system. the agriculture sector and plot a way forward. We
hope it will create interest among governments,
However, such transformation should be made the private sector – including start-ups – civil
in line with principles of inclusivity, affordability, society organizations, farmers’ groups and
accessibility and collaboration. It is important other stakeholders and help them find ways to
that the smallholders and women who contribute collaborate, ensuring the greater vision of “inclusive
significantly to global food production are not left tech for food security for all”.
The agrifood sector sustains the livelihoods of more by a 62% adoption rate among European farmers,
than 1.23 billion people globally.1 The Food and while in Asia adoption is only 9%. The lack of
Agriculture Organization of the United Nations (FAO) clarity on return on investment (RoI) for technology
estimates that in order to feed the predicted 2050 is one of the main reasons for such low adoption
global population of 9.1 billion, the sector will have rates generally. Similarly, women farmers are at
to produce 70% more food than current levels – risk of being left out of the agritech revolution
and to do so with similar resources. At the same due to ingrained socioeconomic challenges
time, the sector is grappling with the challenges such as limited land-holding rights, even though
of climate change, unequal distribution of food women make up 43% of the global agricultural
across the world’s population, soil degradation and workforce. As agritech scales, it is imperative that
geopolitical disruptions to supply chains. Agritech smallholders and women farmers are included,
services are positioned to be the catalysts of the or the implications for food security and the
next revolution in the agriculture sector and have livelihoods of millions of people in emerging
the potential to help farmers produce more without economies will be affected.
requiring increased resources.
Governments and the private sector should explore
In 2021, the World Economic Forum published the kinds of public–private collaborations that have
a community paper as part of the Artificial proved to be effective in creating sustainable public
Intelligence for Agriculture Innovation (AI4AI) infrastructure and services. Agritech services have
initiative that documented more than 20 agritech realized only a small part of the large, untapped
use cases in four categories: intelligent crop opportunity to transform the agriculture sector in
planning; smart farming; farmgate-to-fork; and data emerging economies. Governments will need to
as an enabler.2 This insight report, a sequel to the take on the role of enablers and offer incentives,
2021 paper, takes an overview of the role of Fourth both financial and non-financial, such as access
Industrial Revolution technology in shaping the to quality data through DPI or to on-the-ground
agriculture ecosystem, including farming activities channels in order to encourage the private sector
and supply chains in the same four categories as in to invest in scaling the adoption of agritech in the
the 2021 paper: value chain.
– Intelligent crop planning: gene editing and use It is important to appreciate that digital channels
of artificial intelligence (AI), soil testing-based alone will not drive the adoption of agritech
and sowing-window advice services. There is a need for shared resources – for
instance, rural entrepreneurs, a network of mobile
– Smart farming: AI and augmented reality (AR) money operators or e-governance services – all
for crop advice and field planning, hyperlocal of which could be vital to drive adoption locally.
weather predictions, robotics, yield prediction Since such services would be based in a given
and distributed ledger-based index insurance community, they would enjoy farmers’ trust, and
this in turn would help farmers to understand
– Farmgate-to-fork: traceability, internet of things and appreciate the advantages of agritech and
(IoT)-enabled warehousing, smart logistics and support them in their agritech-adoption journey. A
smart packaging local physical presence would also be important
for the validation of agritech use, confirming the
– Data: digital public infrastructure (DPI – accuracy of satellite data on crops sown in a field,
platform, policy and protocols) and its for example.
implications for farmers
A collaborative effort by governments, the private
As advances in the agritech sector continue, sector – including start-ups and investors –
adoption by smallholders and women farmers, academia and civil society is the need of the hour.
two of the most vulnerable segments, remains a Considering the complexity and interrelatedness of
challenge. As outlined in a McKinsey Insight article,3 activities in the agriculture sector, scaling inclusive
globally only 39% of farmers have adopted at least agritech can best be achieved by taking
one technology service. This average is weighted a multistakeholder approach.
– The availability of agricultural land is not a major A number of agritech solutions have been available
challenge in itself; however, issues such as lack for almost a decade now, and have gone through
of infrastructure and physical inaccessibility to a cycle of research, development, adoption and
markets can make it difficult to access the land scale-up. However, to date these agritech services
that is available. Furthermore, productive land is are yet to achieve scale when compared to the
not equally distributed across all countries, making market potential that emerging economies offer. In
it more problematic for some to produce food this context, the World Economic Forum’s Artificial
locally to feed their own populations.6 Intelligence for Agriculture Innovation (AI4AI) initiative
aims to scale such agritech services through public–
– The economic impacts of natural disasters jumped private partnerships.
Climate change, almost eightfold from the 1970s to the 2010s, from
global geopolitical $49 million to $383 million per day, and human The Forum’s 2021 community paper, published as
activities have increased the probability of more part of its AI4AI initiative, documented more than 20
problems and
frequent heatwaves.7 agritech use cases and technologies and presented
depletion of
a roadmap for these to be scaled in India in four
natural resources – The FAO has reported that 33% of the Earth’s soil broad categories: intelligent crop planning; smart
– especially soil is already degraded and will continue to degrade, farming; farmgate-to-fork; and data as an enabler.9
quality – are the and soil erosion might result in a crop-yield loss of Such categorizations help track the impacts of
issues for the 50%. Human activity such as intensive agriculture, different technologies on the sector and, at a micro
global agriculture deforestation and overgrazing has increased soil level, on farming at a specific location and/or on an
sector now. erosion by a factor of 1,000.8 individual farmer.
Pre-season forecast of Rapid soil analysis and eSHC AI, IoT-driven smart market Agri-data marketplace
demand, supply and prices
Pest prediction and control
Market intelligence Pilot on agri master data
Macro crop planning Controls on application of inputs
at national and state levels
B2B Pilot on agri registries
Smart micro irrigation Aggregation
platform
Dynamic sowing windows
for major crops Drone-based application
Hyperlocal connect Pilot on agri directories
of fertilizers and pesticides
(farmer to consumer)
Plans for priority crops Smart Smart
(offering import-substitution insurance FaaS Farmer to online retail network
options, high value and
high nutrition ) Fintech Uberization Quality
Traceability
assessment
eNWR Digital
Pre-season guidance to input
extension
suppliers, credit and Insurance Smart logistics
Smart CCE, yield prediction (domestic and export markets)
Notes: FaaS = farming as a service; FAIR = Fast Agriculture Interoperability Resources; EFR = electronic farm record; eSHC = Electronic Soil Health Card;
CCE = crop-cutting experiments; eNWR = Electronic Negotiable Warehouse Receipts.
The objective This insight report follows on from the earlier The report also aims to encourage policy-makers,
here is to provide community paper and uses the same four- multilateral organizations, civil society and the
an overview category framework, but it offers a broader global private sector – including start-ups and investors
and agriculture-industry analysis compared with – by providing an overview of the role of agritech
of promising
the farm-focused view of the community paper. currently and where it might go in the coming
technology in
To avoid repetition, the technologies that were years. It is expected that such information will
the agriculture included in detail in the 2021 paper have been help them strategize better and contribute to
ecosystem, the excluded from this report. facilitating the ways in which agritech services can
best ways forward revolutionize the agriculture sector.
and a call to action. The objective here is to provide an overview (and
not an exhaustive list) of promising technology in A team from the Centre for the Fourth Industrial
the agriculture ecosystem, the best ways forward Revolution India, the World Economic Forum
and a call to action to ensure that technology and PwC India carried out consultations with
transformation is inclusive. Agritech innovations are 11 organizations and a desk review to map the
highlighted on the basis of the impact they could game-changing technologies and services that are
have on managing risks from factors including emerging and that could have long-term impacts
climate change, soil degradation, post-harvest on the agriculture sector and its ecosystem. The
losses and demand-price volatility. consultations provided insights on different agritech
– Robotics
– Smart packaging
It is important to highlight that this report is an The objective is to provide an overview and
attempt to map the technologies and not to allow the reader to decide whether to follow up
endorse or promote one technology over another. on any particular area of interest in greater detail.
While there has been a strong focus on increasing The agritech sector is in the process of attempting
production and productivity, the important to create this vision and offer services that
element of reducing cultivation costs by optimizing respond to the needs of the hour and guarantee
resources has not received due attention. future food security.
1.1
Agritech: The situation today
Intelligent crop planning involves the use of AI-based – AI-based soil-testing solutions: Soil testing at
models and other emerging technologies to create the early stages of crop planning is imperative
a detailed, market-oriented and sustainable macro- to ensure better yields. Traditional soil-testing
and micro-level crop plan that also responds to methods may take a few days or a few weeks,
climate-change challenges and the nutritional needs while a sensor and spectroscopy-based soil-
of consumers. Taking into account advances in the testing kit might provide a report in near real
sector, this focuses on: time. However, the soil report itself is not the only
outcome. Technology providers are also adding
– Identifying crops that can withstand the remedial actions based on the soil-test report in
challenges of climate change-induced disasters their modelling, allowing the farmer to take action
or weather-pattern changes while maximizing based on the results.
production and yield and managing risks
– AI tools to predict an optimal sowing period:
– Using AI to predict gene-edited seed performance Changes in weather patterns occur naturally;
depending on different factors however, current weather patterns are attributed
to climate change driven by human activity.10
– Using data to develop models that can Farmers have been adapting to these changes in
predict cropping patterns, harvest periods their own environments – this is especially true for
and production to gauge demand and price smallholders, given how many of them undertake
fluctuations more accurately subsistence agriculture. However, there is a need
to provide farmers with standardized advice
– Targeting efficient nutrient management through and information on ideal sowing windows that is
soil testing based on predictions of weather patterns as well
as the microclimatic conditions at a specific farm
– Accessing digital financial services, including or cluster of farms. Solution providers may use
credit, savings and insurance, to procure quality a range of datasets, including previous weather
inputs, deliver agritech services and manage the data (temperature, precipitation), geospatial
risks in terms of climate change-induced disasters and hyperspectral datasets and microclimatic
data through internet of things (IoT) sensors
Use cases for intelligent crop planning are: and microweather stations to generate advice
for farmers. Studies have shown the benefits of
– The gene editing of seeds using predictive choosing the right sowing window each season
AI tools: Gene editing is where adjustments in in order to maximize yield and minimize climate-
existing genes in plants are made; it is different related risks.
from gene modification, in which external genes
are inserted into an existing genome. The – Augmented reality in crop planning:
objective of gene editing is to drive higher yields Augmented reality (AR) allows farmers to get a
with fewer or similar levels of existing resources. visual feel for the crop plan and layouts of their
These changes are not made by altering just fields by overlaying digital information onto the
one or two genes but are the result of multiple physical environment in which they operate.11 AR-
complex gene edits. AI plays an important role based crop modelling, although in an early phase
in this process. Agritech innovators are using AI of development and adoption, has the potential to
and predictive analysis to analyse plant genes scale if it can be developed as a service that can
and help create a plan for multiple gene edits. be used in an equitable way.
KSA faces arid conditions and water scarcity, challenging traditional farming.
Therefore, there is a strong need to embrace emerging technologies like vertical
farming, controlled environment agriculture (CEA) and precision agriculture with
IoT, and use biotechnology for resilient crops. C4IR KSA is collaborating with C4IR
India through AI4AI for agricultural innovation programmes to bolster food security
and enhance sustainability. We look forward to contributing towards collaborating
with global agritech entities to foster knowledge exchange and implement
advanced solutions to enable a more prosperous and sustainable future.
Basma AlBuhairan, Managing Director, Centre for the Fourth
Industrial Revolution, The Kingdom of Saudi Arabia (KSA)
Since its inception, Agerpoint has made use the fruit clusters and model how much volume
of very expensive LIDAR sensors, employing each cluster will give.
thousands of laser beams to create a 3D
representation and a digital twin of the plant, The key steps in the analysis of this cloud
environment or object of focus. The firm then ecosystem are:
uses data engineering and models to extract
information from the sensors. However, recent 1 Create a digital twin of a plant in 3D
technology developments allow sensor stacks in
Android and Apple products to operate on a par 2 xtract accurate physical and spectral plant
E
with earlier-generation LIDAR sensors costing measurements
$100,000, with the LIDAR sensors in today’s
smartphones able to be used to build AR and 3 ollate data from multiple sources, such as
C
projected reality AR. Banking on this technology drones, field observations, etc.
breakthrough, Agerpoint launched a mobile app
named Capture, which farmers can use to capture 4 reate and deploy automated scalable
C
field data as an image. The app then runs ML models with end-to-end ML platforms
algorithms on the images, offering farmers readily
accessible and accurate agronomic insights and 5
rganize, manage and share insights
O
information they would not have had even two and visualizations for sectors such
years ago. These images are also analysed to as supply-chain management,
detect the presence of disease, identify species research-and-development regulators,
and provide morphological indicators that sustainability/net zero organizations,
diagnose the health of crops and pinpoint fruit marketers and farm growers
types to predict yield. The app algorithms count
Data analytics, 2D/3D modelling and ML primarily use the following metrics:
Rwanda faces agricultural challenges, including limited arable land, water scarcity,
a predominant smallholder farming system and susceptibility to climate change. The
integration of emerging technologies like precision farming, drones, IoT and blockchain
in food-supply systems has transformative potential, enhancing efficiency and
transparency. Keen to collaborate with the C4IR Network, we look forward to adopting
digital agriculture best practices, learning from impactful smallholder farmer case studies
and fostering public-private partnerships for sustainable agricultural development.
Joris Cyizere, Strategy Lead, Centre for the Fourth Industrial Revolution, Rwanda
Platform
Technology platform to allow DPI’s impact on farmers
the exchange of data following
the data owner’s consent
India has 15 agro-climatic zones, which makes it a will be shared through a service layer/API,
very diverse agriculture ecosystem.26 The country’s the Unified Farmer Service Interface (UFSI).
federal government and several state governments
are developing agriculture data stacks with the Agricultural Data Exchange (ADeX): ADeX is
goal of transforming public services and providing an open-source and open-standard public good
a range of datasets to the private sector to foster that allows data sharing between data providers
digital tech in the country. Two case studies are and data consumers, mainly the private sector,
Agri Stack, developed by the Indian Ministry including start-ups, to build customized services
of Agriculture and Farmer Welfare, and ADeX, for farmers. The ADeX platform is coupled with the
developed by the state government of Telangana Agriculture Data Management Framework, which
in collaboration with the World Economic Forum lays out policy to enable the government to share
and the Indian Institute of Science. data with the private sector once the farmer has
given consent for it to do so.
Agri Stack: Designed to facilitate the delivery of
agritech and other digital services to farmers by In both Agri Stack and ADeX, providing access
the government, agritech start-ups, the private to finance, advice and markets are three
sector or other institutions, Agri Stack holds three important elements. In terms of interoperability,
primary datasets: farmer identity; geotagged it is envisioned that a start-up or innovator might
farm location; and crops-sown data. Agri obtain a farmer’s personal datasets from Agri
Stack provides a database and registries, data Stack and also access ADeX for other datasets,
standards, policies for data sharing and APIs. Data such as weather, soil and pest-infestation data.
30%
Trends such as consumers’ preferences for however, the figure for farmers in Asia is only
healthier diets, increased smartphone ownership 9%. One key reason for such low adoption is
and high-speed and cheaper internet penetration, the high cost of technology, as cited by 47%
of farmers cited
advances in AI capabilities, developments in of McKinsey’s respondents.
unclear return on
genetic science and water- and soil-management
investment (RoI)
technologies will all positively affect the scale- A critical insight is farmers’ perception of the
as one of the top
up of agritech services to between $46,000 and value of agritech services. In the same McKinsey
three reasons for not
$60,000 million by 2030.27,28,29,30 While adoption article, 30% of farmers cited unclear return on
adopting agritech.
may continue to grow in larger farms, its adoption investment (RoI) as one of the top three reasons
by smallholders in emerging economies may for not adopting agritech. They also shared that
remain challenging. their minimum expected RoI is 3:1. Farmers were
not able to attribute any change in yield or quality
Currently, only a quarter of US farmers use an to agritech alone as production is also affected by
internet-connected device to access data relating external factors such as weather.
to farming, but the numbers are even lower in
emerging economies.31 As reported by McKinsey The key to scaling agritech will be to work out the
and Company, 39% of farmers globally are unit economics for deploying and adopting agritech
currently using or planning to use at least one services for farmers and showcase empirical
technology in the next two years.32 While 62% evidence about RoI at farm level.
of European farmers have adopted technology,
As the use of agritech expands, women – Supporting agritech adoption through capacity-
farmers must not be left behind in the sector’s building and post-sales services via on-the-
transformation journey. Digital inclusion will remain ground agents
difficult until gender roles are redefined in agriculture
and women’s rights of access to information and – Identifying and highlighting risks and harms, and
decision-making are mainstreamed. building capacity to mitigate these in the early
stages of the adoption journey; issues such
as data privacy must be included in capacity-
building as well as product design
Enhance access
2.3
Harnessing the power of public–private
collaboration
Policy support Collaboration between public and private package focused on the entire value chain have
and the facilitation organizations has been an effective tool in driving all slowed agritech’s growth, hampering it from
of business by the scaling of infrastructure and services, affecting reaching its full potential.
governments populations by initiating projects such as the
coupled with building of roads and ports or development of In PPCs, governments can play a critical enabling
financial systems. Policy support and the facilitation role without acting as the procurer. Procuring
private-sector
of business by governments coupled with private- agritech services and then delivering them to
innovation and
sector innovation and investment create long- farmers alone will not help build a sustainable
investment create lasting, self-sustaining ecosystems. agritech market. The need here is for government
long-lasting, incentives – financial and non-financial – that will
self-sustaining There is a strong case for fostering public–private encourage the private sector to invest in scaling
ecosystems. collaboration (PPC) to scale agritech. Although agritech in value chains or geographical regions.
agritech services have been growing swiftly, Such incentives might include: improving the
scale-up is still limited in emerging economies, availability of data through a data exchange;
and smallholders are affected the most by sector initiating an agritech sandbox (held in agriculture
challenges. Ecosystem impediments, such as the universities) to co-create, test and validate agritech
lack of availability of high-quality and usable data, solutions; and the availability of on-the-ground
lack of technical understanding or knowledge of channels, such as self-help groups, cooperatives,
agriculture in start-ups, the high cost of educating government agri extension workers and/or banking
and onboarding individual farmers – especially agents to onboard farmers. Each government
smallholders – and the fact that agritech is usually can identify the incentives based on the agritech
delivered as a point solution rather than as a holistic ecosystem within its individual jurisdiction.
Leveraging physical channels – by, for instance, journey from adoption to active use and, most
supporting rural entrepreneurs or using an existing importantly, continuing to do so once they are
channel of farmer collectives, cooperatives or onboard. A good example is the Saagu Baagu
organized channels, such as mobile money agents project,34 implemented by the government
or e-governance agents – will be crucial in driving of Telangana, whose partner, Digital Green,
the adoption and active use of agritech services. appointed a cadre of local entrepreneurs who
This can be termed “phygital” – integrating reach out to farmers about the project and bring
physical channels to deliver tech solutions to last- them onboard.
mile customers.
– An on-the-ground presence is also required
The importance of this is down to the to validate digital data in order to bolster AI/
following factors: ML models. For example, while government
agencies and start-ups can provide geospatial
– Digital capacity, trust and perceived RoI all play data on crops sown in each farm, their accuracy
critical roles in the adoption and active use of can be measured only through validation that
agritech services in emerging economies. While further strengthens the AI/ML models which
the lack of any physical presence or interaction analyse the geospatial data. In India, Agri Stacks
leads to lower trust in a new service by farmers, is employing on-the-ground resources to capture
the presence of entrepreneurs/agents from data on the crops sown each season. The plan
the farmers’ own communities helps to build is to share this data with both the government
farmers’ capacity to appreciate the advantages and the private sector’s geospatial data platforms
of agritech services, supporting them on their for validation and to build ML accuracy.
It is clear that agritech powered by Fourth Industrial The World Economic Forum’s agritech initiative,
Revolution technologies will drive the next wave of Artificial Intelligence for Agriculture Innovation,
transformation in the agriculture sector. Challenges is aimed at scaling Fourth Industrial Revolution
such as climate change, unstable geopolitical technologies through public–private partnership.
situations and disrupted supply chains will affect The Saagu Baagu pilot project conducted in
the entire agriculture industry, regardless of the collaboration with the government of Telangana
size of an operation. The advent of agritech focused on introducing four agritech services to
technologies must therefore be viewed from a 7,000 chilli farmers. This saw a remarkable 18%
macro-industry perspective in order to drive value increase in farmers’ absolute profits over their peers
along the entire supply chain, which will help build who were not part of the intervention and is now
resilience, agility, efficiency and transparency for the being scaled to 500,000 farmers planting five crops
agriculture sector. in 10 districts of the state.
It is essential that policy-makers view agritech This initiative highlights the importance of
integration from a systems perspective and multistakeholder collaboration and of a multifaceted
develop digital public infrastructure, policies and approach. The agriculture sector needs similar
programmes to increase the affordability of agritech. collaborative efforts to be targeted at the common
For the private sector – including investors and goal of empowering farmers to adapt to climate
start-ups – and think tanks, efforts to map empirical change and other risks while developing the
evidence of the impact of emerging technologies on industry ecosystem.
farming – yields, quality, revenue, cost of cultivation,
logistics costs and post-harvest crop loss – will Technology is simply an enabler, however. It is
help to generate interest from farmers as well as not a magic wand. It is those involved in agriculture
from small and medium agricultural enterprises, who must empower themselves and take the right
encouraging them to adopt and actively use Fourth decisions by adopting rational methods and making
Industrial Revolution technologies. use of the tools of technological innovation.
Abhay Pareek
Project Lead, Fourth Industrial Revolution for
Agriculture, Centre for Fourth Industrial Revolution,
India
Devansh Pathak
Specialist, Strategy and Community Engagement,
Centre for Fourth Industrial Revolution, India
Acknowledgements