Content uploaded by Luise Pufahl
Author content
All content in this area was uploaded by Luise Pufahl on Mar 17, 2020
Content may be subject to copyright.
Enabling Financing in Agricultural Supply Chains
through Blockchain
1
Inter-organizational Process Innovation through Blockchain
Luise Pufahl, Software & Business Engineering, Technische Universität Berlin,
luise.pufahl@tu-berlin.de
Bridie Ohlsson, Geora, bridie.ohlsson@geora.io
Ingo Weber, Software & Business Engineering, Technische Universität Berlin,
ingo.weber@tu-berlin.de
Garrett Harper, AgriDigital, garrett.harper@agridigital.io
Emma Weston, AgriDigital, emma.weston@agridigital.io
Abstract
(a) Situation faced: One large set of problems in Australian agricultural supply
chains comprises liquidity and payment certainty: farmers are typically paid
for their goods with long delays, and sometimes payment fails altogether.
At the same time, first buyers of the crop have limited funds; and they as
well as the farmers have little access to financing, such as obtaining a loan
against a quantity of the harvested crops. Financiers are often unwilling or
unable to offer financing against such assets, because of insufficient visibil-
ity (or high cost to obtain it). These liquidity and financing shortages are
harmful to the Australian agricultural industry.
(b) Action taken: The AgriDigital founding team knew these issues first hand,
and early on saw the opportunities created by blockchain technology for
their industry, founding the company to solve these key challenges using
emerging technologies. The main hypothesis was that digital trust could be
generated across agri-supply chains using blockchain technology to create
secure digital assets representing harvested crops. This would then become
the foundation data layer to facilitate trade and finance solutions. Effectively
conducting process innovation across the whole industry, AgriDigital first
ran three pilots for understanding the needs and requirements of the different
stakeholders.
(c) Results achieved: Based on these pilots, two outcomes were implemented:
(i) the blockchain part of the technology was spun out into a foundation,
1
This chapter is only a pre-print version. The final chapter will be published by
Springer in Jan vom Brocke, Jan Mendling, Michael Rosemann (Eds.): Business
Process Management Cases, Vol. 2, 2020.
2
Geora, and generalized for other industries, and (ii) a final product was de-
veloped by AgriDigital, using Geora’s blockchain technology, for exposing
a registry for assets, for which financiers can delivery finance in a secure
and streamlined way. In the second half of 2019, already 75 growers and 6
buyers are using the combined solution, from which 122 K bales of cotton
and 34 K tons of grain were financed, with a total value of $18.5 M US$.
(d) Lessons learned: The key learning from running the pilots was that the
emerging blockchain technology can help to improve trust, asset manage-
ment and payments in agri-supply chains. However, applying blockchain
technology needs specific, detailed expertise, and therefore a digital toolkit
that makes the benefits of blockchain accessible on a higher level is of real
value. Further, confidentiality and flexibility requirements of the stakehold-
ers need to be considered. Next to these technological lessons learned also
market-specific aspects are relevant: good behavior should be incentivized;
supply chain stakeholders have low technical capabilities, and on-chain pay-
ments are still a challenge.
1. Introduction
Global agricultural industries produce the food, feed, fiber and many other prod-
ucts that are essential for feeding and clothing the world’s population. In 2018, these
combined industries had a total production value of $3.4 trillion (World Bank 2019).
In a globalized world, the production, processing and distribution methods of agri-
commodities, such as crop and livestock, have created long and complex supply
chains with networks of farmers, processors, traders, logistics providers, financiers,
consumers, and many others; each with their own, sometimes competing, interests.
In a typical agricultural supply chain, farmers provide their harvested crop to a
first buyer (exporters, traders, processors, feedlots), but do not receive payment un-
til a much later date. Commodity buyers usually have limited funds, such that they
need to get additional financing before being able to pay the farmers -- which can
take weeks to months. Thus, the farmer has to burden different challenges, such as
1) lack of liquidity and 2) the counterparty risk (that the counterparty is not paying
at all). Finance for the time between the delivery of the assets and payment would
support farmers tremendously in their business, providing them with the liquidity
to make better marketing decisions or to purchase additional inputs to value add and
improve their products. However, financiers are often unwilling or unable to offer
financing against these assets, because they have insufficient visibility on such as-
sets, or obtaining sufficient visibility is too expensive for the volume of a loan. The
impact of this financing glut is enormous; farmers are unable to invest in new pro-
duction equipment and methods that would make them more sustainable by creating
efficiencies, supporting a transition to better farming practices and boosting overall
yield. The total investment required to reach the respective United Nations sustain-
ability goals is estimated by the FAO to be US$2.1 trillion annually (FAO 2017).
3
The founders of AgriDigital, an Australian-based startup company, experienced
the challenges as grain farmers personally and had further insights into the complete
supply chain from previous businesses. They observed and understood that block-
chain technology can help to improve the trust in grain assets and support payment
security. Blockchain and distributed ledger technologies allow immutable storage
of relevant supply chain information independently from any central authority (Xu
et al., 2019).
Within a year of operation, AgriDigital developed two distinct workstreams (1)
a commodity management application designed to assist farmers, buyers and site
operators with their contracts, deliveries, payments, inventory management and ex-
port documentation, and (2) a blockchain pilot program testing the application of
blockchain across the Australian grains industry. Geora was founded out of
AgriDigital to allow the system designed and tested as part of the blockchain pilot
program, to be rolled out as open-source digital infrastructure for all global agri-
supply chains. Geora developed a digital platform to improve trust between supply
chain participants, provide secure asset registries and facilitate payment and finance
transactions based on blockchain technology.
The process of getting to a production blockchain system followed the green -
fields approach. First, AgriDigital ran three distinct pilots to explore the needs and
requirements of the different stakeholders: farmers, buyers. and financiers; and to
build an understanding of how each participant could benefit from blockchain tech-
nology. Based on the generated insights, a successful new product was developed,
Geora. By architecting a blockchain solution specifically to meet the needs of agri-
supply chain participants, Geora was built to bridge a gap between the user capabil-
ities and requirements, and the core technical components of blockchain technol-
ogy. Together with AgriDigital, the Geora toolkit supports now cotton and grain
farmers with an asset registry and secure payment and provides buyers to get access
to financial products.
Business process management (BPM) is a discipline supporting organizations to
reach operational excellence with regards to their business processes. The tradi-
tional BPM lifecycle, which rather focuses on intra-organizational business pro-
cesses, structures typical BPM activities into phases including process identification
and discovery, analysis and redesign, implementation, execution, and monitoring
(Dumas et al. 2018). In this case study, we report on an inter-organizational process
-- a collaboration between supply chain participants -- observed as a typical target
of blockchain applications (Mendling et al. 2018). In the next section we first iden-
tify the inter-organizational process, and subsequently report on the process discov-
ery and analysis, whereby the situation faced by agricultural supply chain partici-
pants is described with a special focus on farmers and their payment uncertainty. In
Section 3, the three pilots as different options of process redesign will be presented.
In Section 4, the process implementation and execution are explained in terms of
the resulting product, the Geora toolkit, and its design as well as its commercial use
in combination with AgriDigital. In Section 5, technological and market-specific
lessons learned are reported from this process innovation initiative of a supply-chain
collaboration.
4
2. Situation faced
Agriculture, which has not broadly benefited from the ‘first wave’ of digital
transformation, is now undergoing a revolution: Emergent technologies, such as the
Internet of Things, Cloud Computing, robotics, and Artificial Intelligence (AI) are
changing farming tremendously (Rose, DC., Chilvers, J., 2018).
However, while the industry is driving forward with new technologies and the
related opportunities, many farms are run by old-school farmers, following tradi-
tional farming practices, lacking internet access and limited touchpoints to the dig-
ital economy. In Australia, the digital inclusion is 8.5 points higher in capital cities
(62.4) than in country areas (53.9) (Thomas et al. 2018), with the northern and west-
ern rural areas in Australia facing even poorer internet access. Globally, agricultural
supply chains are still largely based on paper records and handshake deals. Infor-
mation about assets is not freely available between the participants; causing a dis-
tinct separation of the information flows relating to the physical assets, data, and
finance as shown in Fig.1. The AgriDigital founding team includes Australian farm-
ers and agribusiness professionals with over 80 years of combined experience in the
grain industry. They personally experienced the challenges of low transparency in
agri-supply chains, where the producers have to burden much of the financial risks.
Next, we describe the traditional process of selling grain from their perspective, as
well as the issues associated with it.
Fig. 1 Traditional agri-supply chains, with separate flows for trade,
data, and finance. © 2018 AgriDigital, reprinted with permission.
Traditionally, Australian farmers have seasonal contracts with commodity buy-
ers, such as exporters, traders, processors, feedlots, who either consume, process or
on-sell the grain. After harvest, many grain farmers elect to transport their grain to
the silo of a third-party site operator (or elevator) where it is stored on behalf of the
5
farmer pending sale to a trader or other counterparty. The discovered inter-organi-
zational process is shown as BPMN collaboration diagram in Fig. 2. BPMN collab-
oration diagrams (OMG BPMN, 2011) can be used to visualize different interaction
partners as pools and the information/asset transfers between them as messages. We
abstract from the detailed internal processes of the single pools as they are not rel-
evant for illustrating the collaboration.
Once arriving on site, the truck pulls up at the weighbridge and quality testing
station. Based on the quality and the amount of grain, the farmer receives a ware-
house receipt, and the grain is stored in a silo of the first buyer, where it is blended
and commingled with grain delivered by other farmers. With the warehouse receipt
and the information of the contract to the first buyer, the delivered products are
indirectly titled to the farmer. It is not yet paid and also not separable from the grain
of other farmers. Then, the grain is further transported, usually by ship to a proces-
sor selling its products to a retailer (a third party), who sells them finally to an end
consumer. The grain is not sold as a single asset as delivered by the farmers, but it
is sold as a bulk commodity, increasing the complexity in asset and title transfer.
The farmer has to wait until the first buyer receives financing or payment for the
harvested assets before getting paid, which can take months.
Fig. 2 Collaboration between the farmer and the first buyer without
involving a financier.
The main reason for the delayed payment is that the data flow is disruptive and
manually handled by the single participants. It requires costly back-office reconcil-
iation processes and manual double data entries burdening the supply chain with
additional costs and human errors. Participants have a hard time verifying commod-
ities and matching payments to title and asset transfers. This lack of transparency
leads, on the one hand, to the situation that participants only trust a limited number
of counterparties, and therefore are hesitant to do business with others, leading to
suboptimal market conditions. On the other hand, financiers are not willing to offer
necessary financing to the farmers or stockholders to provide liquidity. Financiers
have no visibility and do not have enough insights on the harvested assets, such that
an investment is of high risk for them. Obtaining the necessary information might
6
cause high effort and cost; for instance, sending a representative of the financier to
a remote side of a stockholder might cost $1000 just for travel expenses and time
spent; if the loan being sought has a volume of $10,000, it is simply not worth it.
With no financial support, farmers have to carry much of the financial risks,
with the following key consequences:
• Lack of liquidity: With delayed payments, farmers have to focus using
the available funds on short-term investments such as the next harvest, but
long-term investments necessary to ensure business stability and growth,
such as to transition to sustainable production methods or adapting to cli-
mate change, might become impossible to stem.
• Counterparty risk: Although, the farmer is still the owner of the delivered
asset, there is a misalignment between the physical possession, the owner-
ship, and the payment. Therefore, the farmer carries the counterparty risk
until the grain is finally sold. Often this counterparty risk is connected with
the processor at the start of the supply chain. The key challenge is that
grain as a bulk commodity provides no clear visibility on the commodity
ownership. Manual data handling systems provide little to no security for
the farmer when the payment fails. Thus, the farmer has to burden the pay-
ment security of the large parts of the supply chain.
3. Action taken
Experiencing these described challenges, AgriDigital established a blockchain
pilot program testing blockchain in agriculture with the vision to bring together the
physical assets, data, and financial flow. The blockchain, a logically centralized data
store that is physically distributed over a network of participants, allows the storage
of immutable data (Xu et al, 2019). Changes to the state of the data store are limited
to appending new data in the form of a transaction, and such changes require veri-
fication by and consensus among the participants of the network. These features of
a blockchain can support the supply chain participants of grain to have a single
source of truth where the information about an asset, title transfers, and correspond-
ing payments are managed, providing better provence and transparency to the grain
commodity. Authorization to read or write data or participate in the consensus of a
blockchain can be restricted (permissioned or private blockchain) or public. For
fine-grained access control, some blockchains allow the creation of channels: a sub-
set of the blockchain participants become members of a channel, and only those can
see transactions that happen within this channel. For commercial confidentiality, it
is common to restrict visibility in enterprise applications of blockchain.
AgriDigital followed a green-field approach, where they wanted to use the po-
tentials of the emerging blockchain technology and wanted to find out how they can
innovate the existing supply chain with this new technology. To this end, they ran
a pilot program for process redesign, where they tested in constrained environments
7
potential and requirements from three different viewpoints: farmers, first buyers,
and financiers. In the following, the three different pilots are presented.
Pilot 1: Payment Security for the Farmer (December 2016)
In December 2016, AgriDigital executed the world’s first settlement of a physi-
cal commodity on a blockchain. The wheat farmer, David Whillock from Geurie
(NSW), delivered 23.24 metric tonnes of wheat to Fletcher International Exports,
an export business in Dubbo (NSW), and he was instantly paid. The payment was
supported by blockchain technology, specifically a private instance of an Ethereum
blockchain was used. As soon as the quality and quantity of wheat was recorded at
the silo in Dubbo, a smart contract was auto-executed. The smart contract valued
the particular wheat delivery against an existing legal contract and verified that
Fletcher had sufficient funds in their digital wallet to pay Whillock, as shown in the
collaboration diagram in Fig. 3 with the blockchain instance as an active partner. If
this was successful and the physical delivery was finished, the title of the grain was
transferred from the farmer to the buyer, and the payment was simultaneously ini-
tiated. The payment was done off-chain using traditional banking methods, such
that the farmer could receive the money in the local currency. Therefore, a message
was sent to the buyer as a bank file for uploading and paying on the same day.
Fig. 3 Collaboration between the farmer and the first buyer with title
transfer and immediate payment with the blockchain technology.
This pilot showed that the risk of a farmer can be reduced by a payment with
delivery and title creation and transfer to the buyer. However, it deliberately ignored
the financial challenges of the first buyer.
Pilot 2: Extended Pilot with a First Buyer (July 2017)
Next, AgriDigital established a partnership with CBH Group, Australia's largest
grain exporter, and together they ran a pilot at the CBH’s subsidiary Blue Lake
8
Milling, an oat processor in Bordertown (South Australia). Focusing now more on
the requirements of the First Buyer, two important aspects needed to be considered:
1. Providing longer payment terms, here 7 days, while providing security over the
assets of the farmers during that period.
2. Ensuring a level of confidentiality between the participants: procedures, prices,
and details of deals of buyers are subject to commercial confidentiality, and
should not be disclosed to all network participants.
For this pilot, AgriDigital created a user interface and used a private Quorum
instance; Quorum is an Ethereum-based blockchain that allows fine-grained access
control through channels. On this blockchain, AgriDigital created the ‘AgriCoin’ as
cryptocurrency to symbolize payments. The AgriCoin was pegged 1:1 to Australian
Dollars. Also on the blockchain, digital titles for delivered grains were created to
symbolize ownership over these commodities.
Fig. 4 Collaboration between the farmer and the first buyer with a 7-
days payment term and title transfer with the help of blockchain
technology (off-chain payments not shown).
As soon as the farmer’s delivery was tested and weighed at CBH’s site, the qual-
ity and quantity of the delivered oats were recorded in the blockchain-based plat-
form. From these inputs, a smart contract on the platform issued a digital title for
the delivered oats to the farmer. Thus, the farmer could prove ownership of the grain
asset, even though it is in physical possession of the first buyer. However, the digital
title gave better visibility of ownership and rights, such as obtaining a loan against
the asset or selling it. The digital title was flagged for payment after 7 business days.
Once that period elapsed, in an atomic transaction two transfers happened simulta-
neously: the title was transferred to the buyer, and the requisite amount of AgriCoins
was transferred to the farmer. At the same time, the payment of AgriCoins was
mimicked using traditional banking methods (through a platform called Sybiz) so
that the farmer received it in Australian dollars.
9
This pilot showed that the challenge of matching delivery to payment can be
solved. It further showed that a trustworthy digital title of grain can reduce the risk
for the farmer, by delaying the title transfer to the moment of payment. However,
7-day payment terms pose liquidity issues to some stockholders. Therefore, a finan-
cier was involved in the third pilot.
Pilot 3: Grain Commodity Financing with a Bank (December 2017)
In this pilot, the purchase and sale of grain commodities on a blockchain were
demonstrated together with Rabobank, the world’s leading agricultural bank, in a
lab environment. Traditionally, Rabobank provides a product to finance structured
inventory. A grain trader, the first buyer, can enter into an agreement with the bank,
under which the bank acts as an entity buying the grain asset from the farmer. The
grain trader has then to purchase the grain back from Rabobank within a specific
period; only then the legal title passes to the trader. Rabobank adopts the role of the
financier in the process. In the lab environment, it was tested how this structured
inventory product by the Rabobank can be supported with the Quorum blockchain.
Fig. 5 Collaboration between the farmer, the first buyer, and finan-
cier with a commodity financing with the help of blockchain tech-
nology (off-chain payments not shown).
As soon as the delivery has taken place, a smart contract auto-executes the digital
title transfer from the farmer to the Rabobank and initiates the payment on the
blockchain and off-chain. At a later point in time, when the trader is ready to sell
10
the product to a third-party, functions of the smart contracts are invoked to settle
the payment between the bank and the trader, and to transfer the title to the trader.
With this pilot in the lab environment, it could be shown that the blockchain can
ease a time-consuming business process of handling an inventory finance product.
Usually, these kinds of arrangements incur substantial back-office costs and are
only provided to traders having a very good reputation. With the blockchain tech-
nology, a trustworthy digital representation accessible to all participants could be
created, transforming grain assets into a relevant investment object. Further, the title
transfer and payment initiation could be streamlined.
The three options of redesigning the traditional agri-supply chain process
showed that the blockchain technology can increase the trust in the grain supply
chain by working with digital representations of assets. The ownership and the pay-
ments concerning a commodity are visible to the involved supply chain partners and
give a clear view of open payments and loans. Furthermore, the digital representa-
tion of the asset in a blockchain offers also the possibility to involve external finan-
ciers by providing them increased visibility on the asset to be financed and stream-
lining their finance processes, such that they are more willing to finance lower-value
assets. Thus, the availability of financing for grain commodities improves. How-
ever, the pilots also showed that the confidentiality of the data is essential to the
participants of the supply chain and that data should be only shared with permitted
partners. Next, we report on the resulting production system.
4. Results achieved
On the basis of the learnings from these pilots, two results were implemented: (i)
the blockchain part of the technology was spun out into a foundation, Geora, and
generalized for other industries; and (ii) a final product was developed by AgriDig-
ital. AgriDigital’s product is essentially an ERP for participants in agri-supply
chains and uses Geora’s blockchain technology in the backend to interact with a
registry of assets. Through this backend, integration with other supply chain partic-
ipants and financiers is established. In this section, we first present the production
design of the Geora toolkit and then describe how it is commercially used today.
4.1 Production System Design
Geora is technically a permitted Ethereum-based blockchain protocol with an
accompanying developer toolkit that allows its users to easily build solutions to
trade, trace, and finance along agri-supply chains. It aims to be the world’s largest
digital registry of agricultural assets. The toolkit consists of a set of smart contracts
aimed at supporting different supply chain workflows and provides a REST API for
simple access to the smart contract layer from off-chain applications. Additionally,
11
confidentiality is ensured through suitable key management and off-chain data stor-
age, e.g., to enhance the digital asset by certificates, etc. The Geora product consists
of the following three relevant layers:
Fig. 6 Conceptual architecture of the Geora toolkit.
• Smart contract layer (on an Ethereum blockchain): It provides a set of
open-source smart contracts, which executes the logic on the private Ethereum
instance. Smart contracts are available for the creation of digital assets records,
their verification, financial creation, etc. Additional smart contracts can be cre-
ated and deployed to this layer.
Geora uses the Ethereum blockchain to power the smart contract layer. A pri-
vate, permitted network contains nodes that execute and verify all transactions.
These nodes are operated by both Geora and its customers, creating a consor-
tium chain. The network is secured using the IBFT2 consensus protocol, which
provides finality and fault tolerance and prevents bad actors from adding incor-
rect data or breaking the rules of the system. With a current inter-block time of
two seconds and a large block size (i.e., a high block gas limit), the network is
capable of processing hundreds of transactions per second.
• Integration layer: It exposes access to the smart contract layer through a sim-
plified REST API, which assumes no blockchain knowledge. Further, to keep
customer information secure, it provides key management functionality to
manage the cryptographic keys of the business users. These keys are tied to
12
customer identity and used to sign and verify actions in the smart contract layer.
Finally, the integration layer provides a fast, queryable database that reflects
data stored in the smart contract layer and exposes it to applications, and stores
encrypted certificates attached to assets in the Ethereum smart contracts.
• Application layer: It contains tools for building rich workflows on the proto-
col, including financial contracts and agreements, and provides applications for
customer use-cases. AgriDigital is the first customer application on top of the
Geora protocol.
Customer confidentiality and flexibility of data permission are built into Geora
at all layers. Geora supports data confidentiality, making asset and workflow data
available to only those with permission, as well as transactional confidentiality,
which obscures a customer’s counterparties and the actions they take within finan-
cial contracts. To achieve these goals, Geora uses a four-pronged confidentiality
solution:
1. Merkle trees secure asset data at the smart contract layer by compressing all
data into a single hash. The protocol can share this hash across all nodes
without revealing any of the constituent data. Through Merkle proofs, work-
flows and contracts can check specific values in the data without revealing
the entire asset.
2. When customers upload certificates, the protocol encrypts the data using a
unique data key per certificate and stores it on IPFS. Users can share and
revoke access to these files using their own private keys via asymmetric en-
cryption.
3. Each user in the system can hide their identity using pseudo-anonymous on-
demand identities. For each action they take, the user can generate a new
identity using a hierarchical deterministic wallet, that cannot be traced back
to their public identity.
4. Within workflows and contracts, state channels hide the details of actions
from non-participants. The channels perform individual steps of a workflow
away from public view and reveal only the final outcome.
4.2 Commercial Use
Geora supports already five different applications, operating trade and finance
solutions across horticulture, livestock, grain, cotton, and essential oil industries in
Australia and Asia.
AgriDigital and Geora have combined their development efforts using a joint
technical stack to support real-time finance applications in the grain and cotton in-
dustries. AgriDigital’s Commercial Funds have provided finance across cotton, cot-
ton-seed, and grain since its launch in 2018. Geora has supported the product suite
through its digital asset registry, allowing the product to scale and embed security
for the participants since May 2019. In Fig. 6, it can be seen that the product is
already used by 75 growers and 6 buyers. 122 K bales of cotton and 34 K tonnes of
13
grain have been already financed, with a total value of 18.5 M US$. Below, we
describe two of the commercial use cases in more detail.
Fig. 6 Number of users, financed assets, and investments with the
AgriDigital Finance Product and Geora as blockchain backend.
Commercial Use Case 1: Cotton Finance
AgriDigital and Geora designed a system to create a data-rich digital record of a
physical cotton bale, which was then shared with a financier. Having a live view of
the asset provides the financier with the oversight needed to offer growers compet-
itive financial products. For cotton farmers, this financing option is critical. It allows
them to delay selling their asset until after it has been ginned, classed and ware-
housed. This frees up liquidity and allows farmers to optimize their marketing de-
cisions in two ways: first, they better understand the value of their asset after it is
classed, and second, they have greater flexibility in decision making around their
preferred counterparty and the timing of the sale.
Geora’s toolkit was used by AgriDigital to create digital records of cotton bales
in real-time as they are ginned, classed and warehoused. Integrating with data pulled
directly from different parties along the supply chain, a robust and shared digital
record of the cotton can be created. This bale data was then read through the registry
by the AgriDigital special purpose vehicle, giving them the information, they re-
quired to deliver invoice financing to the farmer in real-time.
Commercial Use Case 2: Real-time Grain Finance
Following the first use case, the product expanded into the grains market. Trans-
acting and settling on-chain; creating a more efficient and secure supply chain for
farmers and financiers and linking digital and physical worlds is a focus of contin-
ued innovation. AgriDigital finance software allows the team to manage a massive
number of assets without added friction or an increase in operational resources. The
application is built to handle all asset types, which includes discrete (e.g. a cotton
bale) and bulk (e.g. wheat) assets. The application integrates into the Geora digital
asset registry to provide live updates of the assets in question and any changes to
the quality or quantity of inventory. Drawing from this on-chain data source, each
asset is assigned a market value. To ensure price integrity, the AgriDigital finance
software integrates with the following sources: 1) the AgriDigital platform to access
14
live pricing data 2) a bevy of marketplace web sources via an API and 3) pricing
oracles via the Geora platform.
The Geora registry provides a summary of all assets and finance sent through the
system to financiers, providing the AgriDigital finance product with a streamlined
and transparent way of communicating back with investors in the fund. Leveraging
the Geora engine, M2M calculations, and the AgriDigital platform allows for a fi-
nance product that is disruptive in nature. The traditional finance sector in agricul-
ture is primarily driven by PDFs and Excel sheets, often manual processes, a model
being challenged by these new digital methods of real-time finance.
5. Lessons learned
Over the course of developing a commercial product based on blockchain tech-
nology, from the AgriDigital pilot program to the initial commercial use cases with
the Geora toolkit, several lessons were learned. We structure those into technologi-
cal and market-specific lessons.
5.1 Technological Lessons Learned
Making Blockchain Accessible Through a Digital Toolkit
Over the past years of development, one of the clear lessons learned by the in-
dustry is the difficulties in launching and managing commercial blockchain solu-
tions. Commercial applications require confidentiality, scalability, identity manage-
ment solutions and ease of access that demand very deliberate design. The base
blockchain technology itself is quickly changing and can be difficult for many users
to understand and access. Deep expertise is required to design a blockchain solution.
However, more widespread adoption can be achieved by making design decisions
that abstract some of the blockchain components.
By dividing the architecture of this platform into various layers, the blockchain
can be updated and changed while the technology continues to mature without dis-
rupting the user experience through the integration layer. Essentially, the protocol
can be designed to ‘lift and shift’ as better base technologies emerge. Geora has
developed a generalized set of smart contracts, which can be used for supply chains
in various agricultural industries. These are accessed via well-documented devel-
oper APIs making it much easier to launch applications backed by the security and
scale benefits of blockchain technology.
Confidentiality
One of the key findings throughout the pilot program was that agri-supply chain
participants required a high degree of data confidentiality when accessing block-
chain. The identity of counterparties, transaction values, and production processes
15
are all considered the proprietary business knowledge and commercially confiden-
tial. The tension between transparency and commercial confidentiality is common
in enterprise blockchain use cases (Xu et al, 2019). Full transparency across all data
fields was not accepted. However, where data was disclosed the recipient needed to
be able to trust that it was correct and unadulterated. In response to this, Geora fo-
cused development efforts on building out a confidentiality solution. It meant com-
promises to the truly decentralized nature of the protocol but that was a decision
necessary for launching a commercial solution. Feedback from the customers on the
requirements for this solution was essential in making the trade-off design decisions
around confidentiality vs. decentralization and transparency.
Separation of concerns
Separating the AgriDigital workstream from the Geora blockchain solution was
a critical decision to allow the generalization of the blockchain solution towards
different workflows in the supply chain area. The Geora team, therefore, designed
the Geora API and digital toolkit to allow the user complete flexibility in how they
use the product. While AgriDigital is necessarily more focused on providing a SaaS
solution and maintaining data quality for customers, Geora provides full flexibility
to AgriDigital as a user of the protocol. Everything from key management to per-
mission and data record structures is left to the discretion of the application. This
gives Geora the opportunity to scale the protocol and serve many applications for a
wide variety of use cases.
5.2 Market-specific lessons learned
Low Technical Capabilities
The agri-user typically works off a very low technical base. After each pilot and
customer launch, the technical documentation and processes have been redesigned
to continuously improve the simplicity and scalability of processes like onboarding.
The latest launch of the online developer toolkit has abstracted many of the block-
chain elements to cater to the technical know-how of the Geora user, i.e., application
providers.
Customer onboarding to the protocol can get started with a few devices or sys-
tems currently in operation along the supply chain. As benefits start to be seen from
using digital solutions on-site, they can increasingly integrate and build out a more
comprehensive digital solution to access further features of the blockchain.
Incentivize Good Behavior and Prohibit or Reveal Bad Behavior
When building network technology, attention must be paid to the various incen-
tive models for actors interacting with the system. One of the key design principles
is providing incentives for good behavior and prohibiting or at least revealing bad
behavior. The design of the system and the digital asset must meet the realities of
the industry and the various relationships involved. Various tools can be employed
16
to encourage this. One of the most useful tools in enforcing data and transaction
integrity is the use of permission systems. By establishing roles and permissions
within the system, the system itself can ensure that certain actions are prohibited.
For example, an owner of the asset must be the actor that signs a transaction to sell
that asset.
In the same way, separating the creation of a digital asset from the owner (i.e.
following the four-eye principle (Russel 2005)) is important to ensure that assets in
the system reflect what exists in the real world. For example, where a farmer is the
owner of a physical asset, the site operator who records the quality and quantity of
the delivery will be the one to provide the data on whose basis the asset is created
and issued to the digital wallet of the farmer within the blockchain.
Data contribution is an area where attention should be paid to the incentive model
so as to not encourage bad behavior. Rewarding users for data contribution can en-
courage spamming of the network, at the same time enforcement payments for con-
tributing data can prevent the digital assets from becoming data-rich representations
of the physical commodity. For this very reason, in designing Geora a decision was
made to not charge fees for data reads and writes. Smart contracts that incur an
execution fee are those that result in an exchange of value, i.e., financing or pay-
ments.
Maintaining physical and digital parity is not solved by blockchain technology
alone (Lo et al., 2019), however blockchain offers a tool for users to better assess
the source of the data. In reality, a comprehensive digital solution including IoT
devices provides more accurate sources of data and helps bridge between the phys-
ical and digital world. Photographs and on-chain certificate registries provide ways
of verifying assets and building trust in the data quality and authorship making the
overall digital assets more trustworthy and valuable.
Additionally, manual, human data input continues to act as a threat to data integ-
rity. Integrating with platforms and IoT devices, sensors, and machinery, such as
weighbridges and quality testing instruments wherever possible, reduces the need
for manual data entry. Removing human data input and increasing the number of
such integrations allows for higher data reliability and increases the integrity of the
data stored on the blockchain.
On-chain Payments are Still a Challenge
While the Geora protocol itself can be designed to meet many of the user require-
ments as a commercial solution for asset registries, providing on-chain payment
facilities is still commercially limiting. Cryptocurrencies, bank-backed blockchain-
based currencies, and other blockchain-based solutions like stablecoins have still
not met a point of commercial viability for the agriculture use case. The result of
this is that users of the Geora protocol need to rely on off-chain payment mecha-
nisms for settlement. This is best done through integration with third-party solu-
tions. However, a level of risk is reintroduced by moving onto traditional payment
rails, as it breaks the counterparty security offered by using atomic swaps to execute
17
payments on-chain. Geora’s focus through 2020 and 2021 is on this on-chain alter-
native finance and decentralized finance (defi) space, looking to make it widely ac-
cessible for agricultural supply chain participants.
In conclusion, blockchain technology can change inter-organizational processes
tremendously by providing better visibility on valuable assets to the supply chain
participants. Here, in this case study, it allowed farmers to get a faster payment by
allowing better financing options. Still, the application of the blockchain technology
is still connected with challenges because it requires a specific skill set and the con-
fidentiality of business data has to be preserved.
References
Dumas, M. La Rosa, M. Mendling, J. Reijers, H.A. 2018. Fundamentals of Business
Process Management, Second Edition. Springer.
Food and Agriculture Organization of the United Nations, 2017. The Future of Food
and Agriculture, Trends and Challenges. Rome
Lo, S. K., Xu, X., Wang, C., Weber, I., Rimba, P., Lu, Q., & Staples, M. 2019.
Digital-Physical Parity for Food Fraud Detection. International Conference
on Blockchain, pp. 65-79. Springer, Cham.
Mendling, J., et al., 2018. Blockchains for business process management-challenges
and opportunities. ACM Transactions on Management Information Systems
(TMIS) 9.1, pp. 1-16.
Rose, DC., Chilvers, J., 2018. Agriculture 4.0: Broadening Responsible Innovation
in an Era of Smart Farming. Front. Sustain. Food Syst. 2:87.
Russell, N., van der Aalst, W. M., Ter Hofstede, A. H., & Edmond, D., 2005:
Workflow resource patterns: Identification, representation and tool support.
International Conference on Advanced Information Systems Engineering, pp.
216-232. Springer, Berlin, Heidelberg.
OMG, BPMN ("Business Process Model And Notation Specification Version 2.0),
2011. [Online]. Available: https://www.omg.org/spec/BPMN/2.0/.
Thomas, J., Barraket, J., Ewing, S., MacDonald, T., Mundell, M., & Tucker, J.,
2018: Measuring Australia’s Digital Divide: The Australian Digital Inclusion
Index 2018, RMIT University, Melbourne, for Telstra.
World Bank 2019. World Bank national accounts data, and OECD National
Accounts data files: Agriculture, forestry, and fishing, value added (current
US$): https://data.worldbank.org/indicator/NV.AGR.TOTL.CD (accessed at
30/1/2019)
