Professional Documents
Culture Documents
WEF Artificial Intelligence For Agriculture Innovation 2021
WEF Artificial Intelligence For Agriculture Innovation 2021
Agriculture Innovation
COMMUNIT Y PAPER
MARCH 2021
Cover: Getty Images
Contents
Preface 3
Executive summary 5
3.4 Farmgate-to-Fork 20
Glossary 29
Contributors 30
Endnotes 31
Preface
Emerging technologies from drones to
digitalization have the potential to transform
farming productivity, reduce environmental
impact and boost farmers’ incomes.
Ajay Sawhney
Mr Amitabh Kant
Secretary - Ministry of
CEO, NITI Aayog,
Electronics & IT (MeitY),
Govt. of India
Govt. of India
Ajay Sawhney
Secretary - Ministry of
Electronics & IT (MeitY),
Govt. of India
Jeremy Jurgens
Managing Director,
World Economic Forum
Jeremy Jurgens
Managing Director,
World Economic Forum
Agriculture plays a pivotal role in India’s economy, and exploit the opportunities and challenges of
with over 58% of rural households depending applying upcoming technologies to transform the
on it as their principal means of livelihood. The agricultural landscape, in a way that is profitable
vision of the Artificial Intelligence for Agriculture and sustainable for farmers.
Innovation (AI4AI) initiative is a commitment to
improve the state of the farmers’ world, with the Capitalizing on the value of emerging technologies
operating principle to “think big, start small and such as artificial intelligence (AI), blockchain, drones
scale fast”. The initiative focuses on strengthening and the internet of things (IoT) has the potential
multi-stakeholder collaborations to analyse to impact productivity and efficiency at all stages
The Artificial Intelligence for Agriculture Innovation The report will also serve as a guideline document
(AI4AI) initiative was launched in August 2020 by for agriculture leaders from other states in India and
the World Economic Forum’s Centre for the Fourth in other countries across the world dealing with
Industrial Revolution India, in active collaboration similar challenges in assessing the opportunities
with the Government of Telangana and support for leveraging emerging technologies. The report
from the Ministry of Agriculture, the National helps address some unanswered questions around
Institution for Transforming India (NITI) Aayog and the application of artificial intelligence (AI), the
the Ministry of Electronics and IT. The initiative internet of things (IoT), blockchain, drones and data
now has 60-plus partners across the agriculture governance in agriculture production systems.
industry and emerging technology ecosystem,
actively collaborating through weekly workshops In addition, this report provides a platform for agri-
from October 2020 to March 2021. industries, start-ups, technology organizations,
telecom and cloud service providers, academia and
The initiative’s multi-stakeholder community research institutions to understand the complete
partners have completed a comprehensive review landscape and provides approaches to engage via
of emerging technology use cases, mapping collaborative frameworks.
them against a matrix of viability and impact, and
classifying them as game changers, long-term Considering the expanse and complexity of
interventions and easy wins. This community agriculture production systems and the burgeoning
report is a summary of their findings. It paves the landscape of emerging technology innovations, an
way for the next stage of the initiative, which is intensive and focused effort is required to address
to launch pilot projects and to compile evidence- the challenges and opportunities raised. This
based frameworks through active learning from process needs multi-stakeholder groups of experts
the field. with specialized knowledge and experience in the
areas relevant to AI4AI.
1 Introduction:
challenges and
opportunities in
agriculture
Food and agriculture systems today are – Lack of datasets at farm, farmer and sector
unsustainable for both people and planet. They levels, leading to higher costs of services;
operate at a high environmental cost, waste large
amounts of product and leave many producers in – Gaps in market linkages, challenges in price
emerging markets at or below the poverty level. discovery for farmers and price volatility in
Stakeholders from all sectors and regions have the market;
recognized the urgent need for a fundamental
transformation of food and agriculture systems. – Lack of food processing, logistics and
Such a transformation would create sustained warehousing infrastructure close to farm gates,
social value and deliver greater equity to the increasing wastage;
most disenfranchised.
– Challenges in financial and digital inclusivity;
Agriculture is a high-priority sector of the Indian
86%
economy, with 58% of all families dependent on it, – Poor farm mechanization due to affordability
directly or indirectly, for their livelihoods. The sector challenges.
is at a critical juncture, with many challenges across
of farmers own the value chain. Though the adoption of digital Emerging technologies driven by the fourth
less than two technologies can contribute to addressing some industrial revolution, such as the internet of
hectares of the challenges, success in this area has been things (IoT), artificial intelligence (AI), machine
isolated and has not scaled up adequately. learning (ML), big data, drones and blockchain,
are disrupting many industries, bringing rapid and
Challenges faced by the agriculture sector include large-scale change. Until now, the agriculture sector
the following: has been slow to harness the power of these
technologies. Low adoption levels of emerging
– Small and marginal farmers (86% of farmers) technologies in agriculture are due in large part to
own less than two hectares, causing the complexity of the sector, which features small
unsustainable farm incomes and poverty; farm sizes, lack of telecoms infrastructure in rural
areas, high regulatory burdens which raise costs,
– Unsustainable farming practices, resulting in soil and revenues constrained by customers’ limited
degradation and water stress; ability and willingness to pay.
Agriculture systems in emerging economies – Financial and digital inclusivity for smallholder
are riddled with inefficiency and ineffectiveness. farmers;
Fourth industrial revolution (4IR) technologies are
making it easier to tackle some of these issues. – Safety and trust in food systems;
Improvements in the performance and cost of
computing power, storage and bandwidth have – Real-time monitoring of environment impact;
led to the growth of digital technologies. Taken
together, these technologies can accelerate – Improving productivity and efficiency through
innovation, lower the cost to scale up solutions, farm mechanization;
increase transparency, enable consumers to make
informed choices, and promote evidence-based – Managing food wastage through digital
policy-making. They are also transforming how enablement of storage and logistics
innovations are being conceptualized, designed and infrastructure;
commercialized, and radically changing the ways
businesses operate. – Smart risk management of nature-related
uncertainties.
Emerging technologies, such as AI, blockchain,
drones, IoT and big data analytics, have the potential Emerging technologies, however, are likely to
to enhance productivity and efficiency at all stages of introduce new challenges. They can create
the agricultural value chain, boosting farmers’ incomes, unintended consequences, including data
increasing farm productivity while reducing waste, and breaches and lack of digital inclusivity, which must
enhancing supply-chain efficiency, transparency and be considered and explored in advance. Their
sustainable resource use (see Figure 1). beneficial effects may be unevenly distributed,
potentially deepening the divide between rich
Key objectives of emerging technology interventions and poor. Harnessing the positive impacts of
in agriculture include the following: technological innovation and avoiding potential
downfalls will require deliberate and coordinated
– Sustainable farm incomes through better efforts by investors, innovators and policy-makers.
connected farm-to-market supply chains;
Weather
forecasts
Farm management Land-ownership record
the agriculture
Internet & broadband Mobile ecosystem Big data Internet
of things
Drones
Fertilizers Water
management Precision/smart farming Access to Finance
Source: PwC
Achieving a true transformation of agriculture systems provide growth capital and enable entrepreneurs.
requires a holistic approach, leveraging advancement Civil society will bring the much-needed human touch
in emerging technology innovations and engaging all to ensure the sustainability of these interventions.
stakeholders to accelerate the adoption.
An expansive sector such as agriculture requires a
Every stakeholder has an important role to play. multi-dimensional approach to addressing issues and
Governments can deliver infrastructure and generating impactful, actionable recommendations.
innovative policy, supported by the right financing Against this backdrop, in August 2020 the Centre for
mechanisms. Start-ups can enable innovative the Fourth Industrial Revolution (C4IR) India launched
solutions, by leveraging fourth industrial revolution AI4AI (Artificial Intelligence for Agriculture Innovation),
technology. Established companies can collaborate an initiative to leverage AI and other emerging
to open new markets through sharing data and technologies to make a significant impact on the
intellectual property. Investors and donors can agriculture sector.
The vision AI4AI is conceived as an umbrella programme that innovation, it recognizes that it is also essential to
of AI4AI is “to seeks to create value across the entire agriculture invest in low-tech interventions, create new and
transform the ecosystem through the deployment of a wide range bold policies, improve resource efficiency, build
state of agriculture of emerging technologies, in a way that can scale trust and transparency, align towards common
up across India and provide insights and models for objectives and collaborate across independent
by deploying
other emerging economies. working groups.
emerging
technologies in The vision of AI4AI is “to transform the state of The key objectives of AI4AI are:
an inclusive and agriculture by deploying emerging technologies in
sustainable way.” an inclusive and sustainable way.” – to enhance digital and financial inclusivity among
smallholder farmers;
The AI4AI initiative aims to identify the role of
emerging technologies with the potential for – to build trust and transparency through quality
achieving global aspirations for agriculture systems. and traceability;
It seeks to evolve scalable frameworks for enabling
the use of these technologies for constructive – to protect the environment from unsustainable
outcomes, while underlining the challenges practices; and
and unintended consequences they may bring.
While the programme focuses on technological – to establish sustainable farm incomes.
– Intelligent Crop Planning: working group A In addition, WG members drawn from industry
are representative of the wider value chain of
– Smart Farming: working group B sectors that impact the ecosystems of agricultural
Months
• Define high-level outcomes for each theme
1 • Identify the major opportunities and challenges on holistic basis
• Establish a roadmap to future state with clear next steps and timelines:
From November 2020 to February 2021, C4IR India The Centre has created a steering committee to
helped bring all actors together to align on common coordinate the working groups and their interactions
objectives and enable innovation. Four months of with the wider programme. This multi-stakeholder
continuous dialogue and collaboration revealed advisory body has the following tasks:
the potential impacts of specific technologies and
ways to harness them for positive effect in India’s 1. Provide overall guidance, support and direction
agriculture system. Key findings from this phase of to the working groups to help them operate
work are detailed in the next section of this report. effectively towards achieving their aims.
The AI4AI programme has also launched pilot 2. Coordinate activities across WGs to encourage
projects in five districts of Telangana state, focused synergies and convergence between them and
on priorities in the four defined thematic areas: to avoid duplication of effort.
– Intelligent Crop Planning and sowing – “sow the 3. Validate and approve the frameworks and
right crop in the right soil at the right time”. action plans designed by the WGs in their
respective areas.
– Smart Farming – “maximize productivity,
minimize input costs”. 4. Explore synergies between the AI4AI
programme and other national initiatives.
– Farmgate-to-Fork – “sell produce at the right
time to the right buyer at the right price”. 5. In partnership with C4IR India and the World
Economic Forum, explore synergies and
– Data-driven Agriculture – data governance and convergence with other projects addressing the
data regulation for smart decisions. UN Sustainable Development Goals (SDGs), to
meet local challenges and opportunities, and to
promote investments.
India is ranked second globally for the number and value chains. They prioritized interventions based
of registered “agtech” start-ups. India has 3,116 on two sets of criteria (see Figure 3):
registered start-ups in food and agriculture, and
there has been 25-30% growth in this number – Potential impact: measurable and
year-on-year. Since 2014, $500 million has been non-measurable benefits for the larger
invested in this area and rapid uptake in the last community, analysis of cost-benefit,
12 months highlights the impact of emerging affordability for user groups.
technologies in a post-Covid world.
– Viability: technological viability, operational
The four working groups analysed 70-plus start-up use viability, timeframe of implementation, potential
cases2 across more than 20 themes, crop segments risks and output-action plan.
High
Game changers
These interventions are high impact & high viability –
they could deliver a transformative shift in farming
Easy wins
These interventions are based on increasing farmer
market power and can easily be scaled by supporting
the formation and growth of FPOs
Potential impact
Long-term bets
These tend to be difficult to initiate but reap
significant returns if implemented
Nice-to-haves
These interventions should be carried out in conjunction
with other initiatives and will reap incremental benefits
Low
Low Viability High
This analysis against a matrix of potential impact After 16 weeks of extensive discussions, the
and viability enabled working groups to identify AI4AI multi-stakeholder groups identified a
specific value propositions for each stakeholder set of 11 frameworks and 24 use cases or
group’s set of proposed use cases. The analysis pilots, to be explored further through evidence-
highlighted ways to leverage emerging technologies based learning. Detailed findings in each of
and established roadmaps to achieve results. the four thematic areas are set out below,
The working groups analysed existing pilots while Figure 4 provides a summary.
undertaken by industry elsewhere in India as well as
in Telangana state, leveraging the resources of the
state’s agriculture university.
Frameworks
Varietal advisories Mechanization of farms Quality and traceability DMP for agri-datasets
Extension planning Smart micro-irrigation Supply chain optimization FAIR standard
Credit planning eSHC - rapid soil analysis Fintech, eNWR Master data
Macro irrigation planning Pest prediction Warehousing supply cold chain Registries & directories
Policy/incentive planning Hyperlocal weather advisories Market linkage demand and
price prediction
Inputs planning Smart insurance
Sowing windows, sowing potential, Yield prediction
sowing progress
Crop input advisory and e-marketplace
The Intelligent Crop Planning working group was export), to maximize farm production and
convened with the aim of increasing farmers’ productivity.
profitability and addressing the challenges of climate
change, by deploying emerging technology-led crop – Develop models to predict demand, supply,
planning models at the macro and micro levels. prices and returns with better precision.
The key objective of this framework is to generate
a macro-crop plan for agriculture and horticultural – Enable optimization of planning processes
crops, considering biophysical resources, crop and logistics of upstream and downstream
spread and crop productivity, market intelligence, industries.
consumer preferences, and infrastructure to make
the crop plan comprehensive, inclusive, market- – Develop models to ensure timely availability of
oriented, scalable and impactful. inputs, credit and insurance services to farmers.
The team is creating a robust methodology for – Promote collective and coordinated action by
developing, implementing and upscaling macro- participants in the agri-value chain.
crop plans in diverse geographies, with variable
agro-climatic characteristics, and validating them for – Build preparedness and resilience to climate
viability, robustness and economics. Intelligent crop change and the emission of greenhouse gases,
planning must become an institutional mechanism by increased use of technologies, thereby
for the synergistic interaction of multiple stakeholders reducing risks and maximizing crop yields.
(public and private sector) for the comprehensive
development of the agriculture sector. – Develop emerging technology-led macro and
micro crop-planning models that become
The crop plan strategy features an array of increasingly intelligent through continuous
objectives as follows: learning, and which could ultimately form the
foundation of agriculture growth.
– Identify and promote technology-enabled
cropping patterns that manage the impact of – Maximize income and profitability to farmers
weather (normal and predicted), soil conditions, through validated, adoptable and scalable
water availability and markets (domestic and technology interventions.
Macro-plans are prepared using the most 2. Identification of intelligent crop planning units
representative, viable units of farmland – intelligent (ICPUs) – the most viable units of farmland.
crop planning units (ICPUs). ICPUs are defined by
four factors: soil characteristics and weather, rural 3. Determining variables for ICPUs.
infrastructure and farm credit, market intelligence
and agriculture policy. These factors are considered 4. Determining data availability for designing
the most logical in deciding a state-, national- or ICPUs, for example:
global-level technology-enabled crop plan.
a. public datasets in government & research
The Intelligent Crop Planning working group’s
approach to macro-planning is based on the b. private datasets
following steps and accompanying flow chart
(see Figure 5). 5. Defining the pilot: four pilots are proposed
across four ICPUs over two crop cycles, under
1. Identification of a “workable planning unit” – both irrigated and rainfed conditions; two pilots
the land management unit (LMU). This is a will use only public sector data and technology,
homogenous land unit having consistent soil while two will use both public and private sector
parameters, landscape conditions, climate and (industry and start-up) data and technology. The
length of growing period (LGP). The LMU is one impacts achieved by leveraging the respective
of the four factors required to identify the ICPU. technologies will be measured and compared.
Rural infrastructure
LMUs (soil & weather) Market intelligence Agriculture policy
& farm credit
Conduct pilot
National/global macro-plan
The Smart Farming working group analysed 20-plus 3. Integrated agri-input management: agri-
use cases across eight themes and prioritized them nutrients, irrigation, fintech linkages.
against the potential impact and viability matrix
shown in Figure 3. The analysis led to the selection The working group conducted a detailed study
of the following three themes: of the nine use cases found within these three
themes and developed a framework for the piloting
1. Farming-as-a-service (FaaS); of emerging-technology innovations for evidence-
based learning. The group’s findings are presented
2. Crop health management: yield forecasting, in Figures 6-8 below.
pest management, smart insurance;
India is the leading Nearly 900,000 tonnes of Top two crops for which Top consumer of chemical
producer of pesticides agrochemicals are produced pesticides are used: pesticides is Andhra
in Asia annually: over 50% of the Paddy (26%-28%) & Pradesh with 24% of
total production is exported Cotton (18%-20%) total consumption
every year
Negative environmental Overuse of pesticides is High crop losses in India Higher input costs to Poor agriculture inputs
impacts hazardous to life due to pest attacks farmers advisory services
Only 2% of sprayed pesticides 30% of pesticides are Average 14% crop value losses Excessive usage due to fear of Dependence on input dealers
reach their target species hazardous, e.g. pesticide due to pests pests and recovery of at least for advisories
residues in food system input costs
Impacts on aquatic systems: Overuse of pesticides causes Human and AI-based agriculture
rivers, streams and groundwater Ca. 8,000 deaths/yr, diseases insecticide resistance and Indebtedness and farmer suicides input services required
e.g. cancer, birth defects, pest revival
Effects of pesticides on soil endocrine system
micro-organisms
Solution
Artificial intelligence-based early warning system to provide smart farm-based advisories for pests
Recommended advisory on
Calculate pest density Economic threshold application: timing, type and
from pest trap images limit of pests quantity of pesticide
Agriculture accounts for 70% 53% of cultivated land is India’s irrigation Micro-irrigation
of freshwater withdrawals monsoon-dependent efficiency: 38% deployment: 12%
Resource visibility Water budgeting at village Real-time crop stress Water supply monitoring Lack of smart irrigation
level identification infrastructure and practices
Total water availability from Monitor water supply from
surface, ground and soil moisture Currently command area efficiency Identify crop stress based source to farm to ensure crop Lack of smart irrigation adoption,
is up to 80% on water supply, forecast and stress mitigation lack of conservation infrastructure,
Short- and long-range forecasts crop phenology over-exploitation of groundwater
Total water demand analysis, Monitor releases, monitor canal
surplus & deficit regions and Crop stress monitoring, uneven performance, monitor release vs Crop diversification for
balancing budget rainfall distribution (spatially offtake points till tail end, climate resilience
& temporally), fragmented command area irrigation schedule
Water not received as allocated due visibility, forecast vs stress,
to losses (ET, percolation), conjunctive irrigation advisories
use of groundwater, ensuring
equitable distribution of water
Solutions
Remote sensing AI based weather forecasts IoT & crowd-sourcing Drone watershed planning
Remote sensing data Assemble forecast with Soil moisture and piezometer Higher resolution of drone
clubbed with deep-tech AI micro-level grids to ensure sensors and mobile app data survey digital elevation models
to monitor crop sown area data granularity for integrations with ML models will yield accurate results
and crop water stress true-to-ground data
Crop water need – next 10 days Identify areas with severe dry spell Amount of water required
for saving the farm
Prompt payments to farmers 5 stakeholders – farmer, central Non-loanee farmer subscription Details of insured farmers
not happening government, state government, is abysmally low provided by banks not up to date
insurance company, reinsurance
company
Natural calamity event Deviation between Form-2 Average yield discrepancies, Prevented sowing condition
leading to crop loss & co-observed crop-cutting area sown discrepancies,
experiments (CCEs) CCEs conducted improperly
Crop acreage
estimates
Vegetation index
Maximum VI
Harvest progression,
yield estimates
Full season
crop analytics
Minimum VI
Day of year
The Farmgate-to-Fork working group was Subgroups collectively assessed the relevance of
convened with the aim of addressing challenges each technology solution to their crop segments,
in post-harvest value chain operations and and prioritized them against the potential impact
recommending technological solutions to improve and viability matrix in Figure 3. Based on these
the income of farmers and improve returns for all prioritizations, subgroups assigned each technology
supply chain players in the agriculture ecosystem. solution one of the following four categories:
The working group was divided into four crop Game changers: High potential for farmers and
segment subgroups – cereals and pulses, fruits, overall value chain impact, with strong technical and
vegetables, and plantations, spices and oilseeds – commercial viability.
with each crop segment having common producer
and market characteristics. Subgroups were Easy wins: Medium potential for farmers and
allocated participants from the working group who overall value chain impact, with strong technical and
were known to have technical and commercial commercial viability.
experience in that crop segment. Each subgroup
assessed and developed case studies of 20- Long-term bets: High potential for farmers and
plus promising technological solutions. Some overall value chain impact, with medium technical
solutions were sourced from the agri-technology and commercial viability in the immediate term.
compendium developed by the Government of
Telangana, and some solutions were provided by Nice-to-haves: Low to medium potential for
working group members. farmers and overall value chain impact, with low to
medium technical and commercial viability.
Technology solutions were spread across six
functional areas – quality assessment, logistics Figure 9 below details the Farmgate-to-Fork
and warehousing, financial services, buyer-supplier technology solutions, organized by functional area
matching, traceability and market risk management. and priority.
Cereals Plantations,
Functional area(s) Innovation Vegetables Fruits spices,
& pulses
oilseeds
Quality Cloud-based AI for grading and sorting Game changer Nice-to-have Game changer Deprioritize
assessment
Spectral and machine vision-based Easy win Nice-to-have Nice-to-have Game changer
quality assessments
Image processing-based grading & sorting Easy win Long-term bet Long-term bet Nice-to-have
Logistics and Solar-powered on-farm cold room w/ IoT Easy win Nice-to-have Easy win
warehousing
Warehouse management SaaS Nice-to-have Deprioritize Long-term bet
Blockchain-powered warehouse
Easy win Nice-to-have Long-term bet Nice-to-have
receipt financing
Buyer-supplier B2B electronic marketplace Game changer Game changer Game changer Game changer
matching
Farmer-level marketplace Long-term bet Long-term bet Long-term bet Long-term bet
Crop segment Demand centre Buyer-supplier Quality Traceability Logistics & Financial
matching assessment warehousing service
Government
Traditional retailer
Vegetables
Organized retailer
Value-added processor
Traditional retailer
Fruits
Organized retailer
Value-added processor
Among the deep-dives into the challenges and solutions in crop Centric function Other value-added functions
segments and demand centres, buyer-supplier matching as a
centric function is identified as a technology solution common to
8 out of 10 model value chains.
Source: Working group input, TechnoServe India analysis Artificial Intelligence for Agriculture Innovation 21
The complete Following further discussions on individual solutions, – An online retail platform linking retailers and
stack of buyer- the working group concluded that a B2B buyer- consumers would help promote e-grocery
supplier matching supplier matching platform model that facilitates development in India. Some existing online
platforms is trading between aggregators/intermediaries and retail platforms have already demonstrated high
buyers – with integrated services in logistics and maturity of functionalities, even though they are
estimated to create
warehousing, quality assessment, traceability and still very limited in scale.
an additional value financial services – presents the highest potential
of $57-70 billion to drive value capture and value creation in the – Hyperlocal retail platforms facilitating direct
over a 20-year ecosystem. Alternative buyer-supplier matching sales between farmers and consumers
period. models are also being conceptualized, for example: would most directly improve value
farmgate aggregation platforms that link farmers and captured by farmers. Existing solutions
aggregators/intermediaries, online retail platforms in hyperlocal retail platforms are still at a
that link retailers and consumers, and hyperlocal nascent stage and need further functionality
platforms that directly link farmers and consumers. improvements and operational refinements.
Other conclusions from the working group included The complete stack of buyer-supplier matching
the following: platforms is estimated to create an additional value
of $57-70 billion over a 20-year period. In this
– The farmer aggregation platform would be best steady state, an annual additional value of $15-18
led by the national government to ensure fair billion would be generated for farmers. Within 20
price capture for smallholder farmers. Existing years, between 61 and 75 million farmers (who
practices in this area have shown low maturity would constitute 35-43% of the farming workforce)
of functionalities and require more investment in would stand to gain from these platforms.
setting up last-mile operations.
Dataset Description
1 Soil health Agronomic details like soil-type and fertility, including nutrient availability
(macro, micro, secondary), moisture content etc. for each farm
2 Satellite imagery High resolution images to identify farm boundaries, crop distribution, yield etc.
3 Real time data on Real time statistics on variety-wise market price and arrivals data from
agriculture markets commodity trades
4 Crop yields Data on actual yields for crop varieties per area harvest
5 Production & Production and consumption volumes for crop varieties by month and location
consumption data
6 Weather data Climate details including rainfall, precipitation, humidity, sunlight, temperature,
wind etc. at district level
7 Irrigation maps High resolution irrigated area mapping to identify areas under irrigation,
moisture levels in top soil, root zone etc.
8 Storage network Storage network details like crop varieties stored, maximum capacity, average
details utilization and safety buffer
9 Warehouse details Warehouse details including locations, facilities like cold storage, capacity
constraints, tariffs, operating and handling costs, fixed costs
10 Commodity Profile including standards for defects based on crop varieties & usage, shelf
profile data life, trade constraints, purchase limits, timing of production
11 Digital land Digital land records registry that establishes collateral and has legal validity with
records registry various departments (revenue, survey etc.)
12 Defect & Annotated dataset of images of different crop varieties for AI-based grading,
pest images diagnosis and defect identification
14 Import, export Import and export volumes for crop varieties by month and location
volume details
15 Historical purchase Historical daily purchase prices for crops by location, market type, level (e.g.
Source: NASSCOM prices for crops farmer, middleman etc.)
and McKinsey
– An environment of trust
FIGURE 13 Data marketplace use cases – pest condition prediction & advisories
nt ecosystem
pest manageme
covery for services in
Data value dis
Data sets Type of data Transformation Data market Data consumer Services
Current pest data (crop System integrators: States, NCDEX Start-ups (advisory): Farmers
Pest data images, remote sensing) Tech M, Wipro Cropin, Satsure,
Wadhwani
Research, universities,
crowd-sourcing, Historic pest Start-ups: Satsure, eNAM Start-ups (agri-input): Government
institutions incidence data Wadhwani Unnati, Vassar, Dehaat,
Agrostar
Note: names of organizations are representative only and not an exhaustive list
4 Conclusion
and next steps
The AI4AI initiative is aimed at demonstrating value 3. Smart FaaS (farming-as-a-service), an ET-based
to the main stakeholder of the agriculture sector, one-stop-shop for farmers’ needs throughout
namely the farmer, through a multi-stakeholder the agri-cycle;
approach. The AI4AI impact objectives for phase 1
are defined as follows: 4. Creating e-SHCs (e-soil health cards) and
making them accessible online to the target
– One goal: the economic development of farmers group of farmers;
through a multi-stakeholder approach
5. Implementing smart crop insurance in the entire
– 10 streams of new value (e.g. seven use cases project area;
across three different themes)
6. Demonstrating traceability of agricultural
– 100 farmer-producer organizations produce in respect of high-value agri-produce;
The programme is to be implemented in a progress of the project and ensure timelines are met
mission-mode, given the seasonality of the and bottlenecks removed at the field level.
agriculture sector. As such, the following multi-level
implementation framework is proposed: A significant effort will be made on capacity
building at multiple levels, especially at the field level,
An empowered committee at the state level, led including the functionaries of agriculture, revenue,
by the principal secretary (agriculture), is proposed development, irrigation and other departments. The
to define the vision, give direction and take all the methods should include workshops, newsletters,
project-related policy decisions. It would consist of knowledge portals and webinars.
representatives of all major stakeholder groups. It
may meet on a monthly basis. An effective awareness campaign should be
launched in project areas, to ensure the increased
A steering committee at the state level, led by uptake and utilization of project services and
a senior officer of the domain, should review and offerings by farmers, FPOs and market players of
guide the project on a weekly basis, and take all the the agri-ecosystem.
operational decisions.
C4IR India shall play a facilitatory role
District-level project committees, headed by the in the smooth implementation of the
concerned district collector, should review the project and realization of the goals.