Title:
SYSTEMS AND METHODS FOR CREATING DYNAMIC CREDIT LIMIT AND RECOURSE BASE FOR SUPPLY CHAIN FINANCE
Document Type and Number:
WIPO Patent Application WO/2022/046407
Kind Code:
A1
Abstract:
Systems and methods of dynamically creating credit limits and automatically mitigating risk in real time provide digital supply chain finance services for buyer-supplier pairs based on sets of related fast data and other event driven applications. Suppliers who sell goods and services to other businesses (buyers) use the systems and methods of the invention to request payment up to the dynamic credit limit from a third-party supply chain financial provider. Dynamic credit limits are calculated automatically based on digital analysis of data sets automatically pulled from multiple sources. Suppliers request payment against an approved and scheduled payment invoice by a buyer ("confirmed invoices") earlier than the scheduled payment date ("early payment"). Supply chain finance providers offer a fully digital early payment solution against confirmed invoices where risks of non-payment or receipt of diluted payments for funded invoices are mitigated by a structured recourse base that is automatically created by establishing the dynamic credit limit.
Inventors:
SHAPIRO GEORGE (US)
Application Number:
PCT/US2021/045483
Publication Date:
March 03, 2022
Filing Date:
August 11, 2021
Export Citation:
Assignee:
IFG NETWORK LLC TRADING AS THE INTERFACE FINANCIAL GROUP (US)
International Classes:
G06Q40/00
Foreign References:
US20080046334A1 | 2008-02-21 | |||
US20140067650A1 | 2014-03-06 | |||
US20180182029A1 | 2018-06-28 | |||
US20150026027A1 | 2015-01-22 |
Attorney, Agent or Firm:
PARISI, Joseph, A. (US)
Download PDF:
Claims:
What is claimed is: 1. A digital supply chain finance system for creating a dynamic credit limit, the system comprising: a dynamic credit limit server application, including instructions stored on a non‐transitory computer‐readable medium executed on the dynamic credit limit server that is configured to receive a request for service from a supplier via a supplier portal, automatically receive confirmed invoices from a buyer enterprise resource planning server, automatically receive fast data from a data source, and determine the dynamic credit limit based on a gross amount of confirmed invoices issued to the buyer from the seller, a base ceiling, a historic dilutions multiplier, and a risk score multiplier. 2. A digital supply chain finance system of claim 1, wherein the dynamic credit limit server is one or more virtual servers running in a cloud computing environment accessed by the supply chain finance system via a data exchange network. 3. A digital supply chain finance system of claim 1, wherein the fast data includes real‐time data received from an event‐driven server application of the data sources. 4. A digital supply chain finance system of claim 1, wherein the fast data is received from an event‐driven server application of the data source via an application program interface (API) or a secure file transfer protocol (SFTP). 5. A digital supply chain finance system of claim 1, wherein the fast data includes actionable data received from at least one of the group of sensors, actuators, and machine‐to‐machine data exchange sources. 6. A digital supply chain finance system of claim 1, wherein the base ceiling includes an initial maximum percentage of the gross amount of confirmed invoices issued to the buyer available for funding by the supply chain finance provider. 7. A digital supply chain finance system of claim 6, wherein the base ceiling is established per buyer by underwriters of the supply chain finance provider. 8. A digital supply chain finance system of claim 1, wherein the historic dilutions multiplier is based on at least one of the set of historical dilution events from an enterprise resource planning server of the buyer and a dilution prediction for the buyer‐supplier pair. 9. A digital supply chain finance system of claim 8, wherein the historical dilution events from the enterprise resource planning system of the buyer include automatic analysis of the data of historic payment amounts for confirmed invoices and corresponding gross amounts for the same confirmed invoices issued by the supplier for a predetermined time period. 10. A digital supply chain finance system of claim 8, wherein the dilution prediction for the buyer‐ supplier pair is based on a machine learning (ML) model trained by billings and correspondent payments from 1st tier buyers to their suppliers in at least one of the group of different industries, different jurisdictions, and different economic cycles. 11. A digital supply chain finance system of claim 1, wherein the risk score multiplier is based on at least one of the set of a financial analysis score, based on at least one of the set of a profit and loss statement, a balance sheet, an accounts receivable aging report, and an accounts payable aging report received from accounting software of the supplier; a fraud‐threat score, wherein the fraud‐threat score is based on at least one of the set of an email risk based on an email address age, an IP address confidence based on a historical IP address fraudulent use, an IP address risk based on a location of an IP address and a supplier location, a Proxy‐ VPN‐TOR determination based on a direct or non‐direct connection, an email free‐corporate determination based on whether a user is using corporate email or free email, a city confidence score based on a location of a city and proximity to the supplier’s address, a geolocation score based on a supplier’s address and IP address, a location accuracy radius score based on a supplier business location and user IP address distance, a location average income score based on a weighted average income per person for a postal code associated with the IP address, a postal confidence score based on a business account address or a credit card address and a user address, and an address‐phone residential‐business score based on a user connection location); a compliance score, wherein the compliance score is based on at least one of the set of an international watchlist score based on an Office of Foreign Assets Control sanctions list, an enhanced credit score based on a combination of data from one or more registered credit agencies and augmented with at least one of the set of utility records, electoral rolls, and drivers’ license records, a passport‐driver license‐ID validation score, an address validation score, and a utility score based on correlation of utility bills with a user address; a business credit score, wherein the business credit score is based on at least one of the set of an active registration‐time in business score, a derogatoriness score, an insolvency history score, a collection‐revenue ratio score, a tax liens‐CCJ history score, and a trade names score, wherein each of the at least one of the set of scores is based on credit reports in data‐feed format pulled from a credit bureau; a filed liens score, wherein the filed liens score is based on at least one of the set of a number of not‐terminated filings (UCC/PPSR/Charges) score, a number of not‐terminated filings (AR/Debtors) score, a not‐terminated filings (Inventory) score, a not‐terminated filings (PMSI) score, a not‐terminated filings (All Assets) score, all based on collateral descriptions, and an alternate payee score based on existence of an active payment assignment to a third party; a success‐social value score, wherein the success‐social value is based on at least one of the set of a level of education score, a career score, a loyalty score based on frequency and duration of job changes , a Social Activity score based on at least one of the set of user social involvement in charities, community activities, and military background., and a reputation score based on collected data related to user‐signer reputation; and a track record score, wherein the track record score is with the supply chain finance provider and is based on at least one of the set of a longevity of account score, a defaults number to transactions number ratio, a ratio of post‐confirmation dilution to gross amount of all confirmed invoices, a ratio of post‐confirmation dilution to a total of an amount of automatic recourse available in the current payment period and an amount of automatic recourse available in the next payment period, and a total confirmed invoices score. 12. A digital supply chain finance system of claim 11, wherein each of the scores from which the risk score multiplier is based, is statically or dynamically weighted based on the received fast data. 13. A digital supply chain finance system of claim 12, wherein a different static or dynamic weight multiplier applies to each score to differentiate contributions of each score to the risk score multiplier. 14. A digital supply chain finance system of claim 11, wherein the financial analysis score is also based on at least one of the set of an extrapolated income determination, an equity‐income ratio, an assets‐liability ratio, a monthly revenue ratio, a total accounts payable‐accounts receivable comparison, and a time‐based accounts payable‐accounts receivable comparison, a highest accounts receivable concentration, and a reporting period. 15. A digital supply chain finance system of claim 11, wherein the risk score multiplier is also based on at least one of the set of an Authentication Result score, an ID Passed/Failed/Unknown score, a Total ID Verification score, an ID Barcode Verification score, and an ID Data Extraction Reliability Level score. 16. A digital supply chain finance system of claim 15, wherein the least one of the set of the Authentication Result score, the ID Passed/Failed/Unknown score, the Total ID Verification score, the ID Barcode Verification score, and the ID Data Extraction Reliability Level score is based on analysis of driver licenses or passports for a predetermined jurisdiction. 17. A digital supply chain finance system of claim 11, wherein the post‐confirmation dilution includes at least one of the set of an amount of unrelated credit‐memos applied on the same date as an invoice scheduled payment date, an amount of chargebacks, withholdings, counterclaims and an amount of set‐offs. 18. A digital supply chain finance system of claim 11, wherein the loyalty score includes a level of continued effort to achieve a goal and a measure of jumping from project to project. |
Description:
SYSTEMS AND METHODS FOR CREATING DYNAMIC CREDIT LIMIT
AND RECOURSE BASE FOR SUPPLY CHAIN FINANCE TECHNICAL FIELD [0001] This technology relates to systems and methods for s
upply chain finance. More particularly, the technology relates to systems and m
ethods for analyzing credit risks and risk of invoic
e payment dilutions for business organizations, determini
ng a dynamic credit limit and a recourse base in real time, and providing early payment services and
solutions against confirmed invoices based on the dynamic credit limit. BACKGROUND [0002] Supply chain finance (SCF) includes technology‐based
solutions to lower financing costs and improve business efficiency for buyers and seller
s linked in a sales transaction. Supply chain fina
nce methodologies work by automating transactions and trac
king invoice approval and settlement processes from initiation to completion. Buyers agree to appr
ove their suppliers' invoices for financing by a ban
k or other outside financier. By providing short‐ter
m credit that optimizes working capital and provides liquidity to both parties, supply chain fina
nce offers advantages to all participants. Suppliers
gain quicker access to money they are owed, and buy
ers get more time to pay off their balances. Both
parties can use the cash on hand for other projects
to keep their respective operations running. smoothly. [0003] Supply chain finance includes "supplier finance," “p
ayables finance,” and "reverse factoring," and encourages collaboration between buyers
and sellers. This philosophically counters the competitive dynamic that typically arises between a b
uyer and seller. Under usual circumstances, buyers attempt to delay payment, while sellers look
to be paid as soon as possible. [0004] Supply chain finance can optimize cash flows and wor
king capital by allowing buyers to lengthen their payment terms to their suppliers while
providing the option for their suppliers to get pa
id early. Supply chain finance is often used when the
buyer has a better credit rating than the seller
and can consequently source capital from a bank or other
financial provider at a lower cost. This advantag
e lets buyers negotiate better terms from the seller,
such as extended payment schedules. Meanwhile, the seller can unload its products more quickly, to
receive immediate payment from the bank or other intermediary financing body. [0005] A traditional supply chain finance situation involves
extending payment terms. An extended payables transaction might include a buyer (
company) that purchases goods from a seller (supplier). Traditionally, supplier ships the goods,
then submits an invoice to company, which approves
the payment on standard credit terms of 30 days, fo
r example. But if supplier is in urgent need of
cash, supplier may request immediate payment, at a discount
, from the company’s affiliated financial institution, such as a bank. The bank or supply c
hain finance company intermediates the accounts receivable process for the buyer. When the immediat
e payment request is granted, that bank or financial institution issues payment to supplier, and
in turn, charges the company a fee and extends the
payment period for company for an additional further
30 days, for example, for a total credit term of
60 days, rather than the 30 days mandated by supplier.
[0006] However, supply chain financing is limited to compani
es that can be credit insured or is made available only to investment grade or near‐inv
estment grade suppliers. If the buyer cannot be credit insured or is sub‐investment grade, the bank
or some supply chain finance companies will not offer this solution. [0007] In addition, supply chain finance providers, such as
banks and other financial institutions throughout the world, are currently mitig
ating buyers’ non‐payment risks and payment dilution risks of invoices payable to suppliers by r
equesting and obtaining buyers’ guarantees of payment for invoices which have been paid early. T
his current approach of addressing risks related to providing supply chain finance service creates sev
eral problems. For example, the guarantee of payment limits buyers’ abilities to apply legitimate
chargebacks, set‐offs, counterclaims, and withholdings to the invoice payments payable to suppl
iers. [0008] Another problem that buyers are facing by providing
a guarantee is an increasing requirement for the modification and de‐recognition
of the buyers’ trade payables and reclassification of trade payables as a debt on the
buyers’ balance sheets. Such requirements can create significant accounting treatment issues for buy
ers when classifying both liabilities to a trade creditor and liabilities to a financial institution a
s debt. [0009] These technological problems create limitations to the
buyers’ ability to initiate and offer supply chain finance service for their supplier
s. Several attempts to create technical solutions that allow waivers of the buyers’ guarantee require
ments for supply chain finance services have been unsuccessful and unreliable. SUMMARY [00010] The systems and methods of the invention provide tec
hnical solutions to the dearth of current supply chain finance methods and systems.
The invention is based on a dynamic credit limit that reliably allows banks and other financial
institutions to offer supply chain finance service without requesting a payment guarantee from a buyer,
which is traditionally used as a hedge against the risks of non‐payment and payment dilution. [00011] Systems and methods of the invention determine a dyn
amic credit limit and a recourse base in real time, and provide early paymen
t services and solutions against confirmed invoices based on the dynamic credit limit. The dy
namic credit limit systems and methods described in this disclosure provide a technological solution t
o an issue rooted in technology, including improved systems and methods for processing and analy
zing disparate data in large volumes at scale from multiple sources. [00012] The invention provides methods and systems for real
time calculation of a dynamic credit limit and related recourse base for a Supplie
r‐Buyer pair, allowing supply chain finance providers to make early payments to Suppliers without
asking Buyers for a payment guarantee while simultaneously reliably mitigating Buyer’s non‐payme
nt and payment dilutions risks. Specifically, the systems and methods of the invention allows analysis,
in real time, of historical payment dilutions between a specific Buyer and Supplier and issuance o
f the credit/risk score, also in real time, based on the analysis of a variety of data, instantly pul
led from thousands of data points in data‐feed formats, related to fraud/threat, credit history, lien
s/taxes/judgments, compliance, and other information that can be used to predict payment dilu
tion and necessary recourse base for the buyer‐ supplier pair. [00013] The dynamic credit limit (supply chain finance provid
er) server may unconventionally utilize data from a variety of third party platforms
, including e‐invoicing platforms, accounting systems, buyer ERP systems, fraud/threat data sources,
credit bureaus, compliance data sources, liens searches sources, social value sources and othe
r platforms from which large data sets may be analyzed computationally to reveal patterns, trends, a
nd associations that provide an analytical information‐based platform for determining a dynamic
credit limit in real time. For example, the dynamic credit limit server may analyze rich data fr
om third party platforms, including data on suppliers and buyers (e.g., a customer of a supplier
), and the dynamic credit limit server can generate
on‐the‐fly determinations of payments for invoices
chosen by the supplier, less a discount. [00014] The dynamic credit limit server analyzes the data an
d provides a dynamic credit limit that allows supply chain finance service without buye
r payment guarantee. By seamlessly integrating the dynamic credit limit server with buyer ERP syste
ms and supplier accounting software (e.g., QuickBooks®, Xero, FreshBooks, Sage, NetSuite, etc.),
supply chain finance providers can accurately provide a dynamic credit limit. Similarly, the inve
ntion extends its use of fast data from third party
platforms into the dynamic credit limit calculus and
can forgo a Buyer payment guaranty to mitigate non‐payment and payment dilution risks. In this f
ashion the invention creates significant improvements in widening supply chain finance services
worldwide. For suppliers, the benefits include a reduction of trade receivables and an incr
ease in cash position, faster access to cash at advantageous rates, and stronger ties and cooperation
with the buying company, which can create competitive advantages. Additionally, suppliers benefi
t from a faster cash conversion cycle from delivery to cash, working capital optimization, and i
mproved liquidity. Benefits for buyers include maximized use of e‐invoicing, self‐billing, and co
operation with suppliers. [00015] Users (suppliers) choose a payment date for invoices
that their customer (buyer) has approved and scheduled for payment in exchange for p
aying a small discount to the invoice value to the supply chain finance provider. The invention al
lows suppliers to view invoices approved and scheduled for payment by their customer (buyer) and
to choose invoices for early payment. As users (suppliers) choose invoices, the systems and methods
of the invention continually calculate and update the dynamic credit limit. Suppliers can take
early payment on as many invoices as desired, up
to the limit of the dynamic credit limit. The sys
tems and methods of the invention simultaneously calculate and update a recourse base as well. [00016] One example of how the dynamic credit limit determin
ation can be used is when suppliers need a specific amount of money (e.g., for
payroll, inventory, working capital, etc.), a cash
planner feature enables a user to enter an amount o
f money needed and automatically identifies invoices for early payment that total to the amount
of money entered by the supplier. This provides a powerful cash management tool. [00017] An additional example of how the dynamic credit limi
t determination can be used is when suppliers would like the digital supply chain f
inance method to be performed automatically (for example, without reviewing a list of invoices for ea
rly payment), the supplier can select an “always pay me early” feature. This feature provides ongo
ing early payment of invoices approved and scheduled for payment by the buyer up to the amount
of the dynamic credit limit. [00018] This innovation is creating a universally accessible
system by transforming the score‐ related information about the Buyer‐Supplier pair in
real time into a factual machine algorithm which can be easily used by any supply chain finance serv
ice providers. [00019] Prior supply chain finance systems and methods are n
ot designed with a 360‐degree view of Buyer‐Supplier data in real time, in combi
nation with the pair’s historical payment dilution
data analysis, and thus, these prior systems are una
ble to offer a reliable risk mitigation solution. Those methods are also more prone to error because
they are unable to automatically take into account the vast majority of risk factors that are
critical in helping to determine a proper risk mitigation structure. [00020] This invention provides systems and methods of automa
tically creating a dynamic credit limit at the point‐of‐funding decision base
d on an analysis of critical risk factors data obta
ined in real time. The systems and methods of the inve
ntion, in turn, automatically structure a recourse base, allowing supply chain finance providers to miti
gate non‐payment and dilution risks. These systems and methods allow supply chain finance funder
s to provide supply chain finance service reliably and safely without requiring a buyer’s pay
ment guarantee. [00021] Systems and methods in accordance with the invention
include a digital supply chain finance system for creating a dynamic credit limit.
In one example embodiment, the system includes a dynamic credit limit server application, including
instructions stored on a non‐transitory computer‐ readable medium executed on the dynamic credit limit
server. The dynamic credit limit server and server application are configured to receive a reques
t for service from a supplier via a supplier portal
. The server and application are further configured to
automatically receive confirmed invoices from a buyer enterprise resource planning server, and to aut
omatically receive fast data from a data source. Additionally, the servers and applications are configu
red to determine the dynamic credit limit based on a gross amount of confirmed invoices issued to t
he buyer from the seller, a base ceiling, a histori
c dilutions multiplier, and a risk score multiplier. [00022] Some embodiments of the invention are instantiated in
a cloud server environment, where the dynamic credit limit server is a virtual
server running in a cloud computing environment accessed by the supply chain finance system via a d
ata exchange network. [00023] The fast data received can be real‐time data recei
ved from an event‐driven server application of the data source and can be received
from the event driven server application via an application program interface (API). In some instanc
es, the fast data includes actionable data received from sensors, actuators, and/or machine‐to
machine data exchange sources. [00024] Some examples of the invention include a digital sup
ply chain finance system where the base ceiling includes an initial maximum percenta
ge of the gross amount of confirmed invoices issued to the buyer that are available for funding
by the supply chain finance provider. The base ceiling can be established per buyer by underwriters
of the supply chain finance provider. [00025] Systems and methods in accordance with the invention
can establish scoring multipliers in a number of ways. For example, the
historic dilutions multiplier can be based on historical dilution events from an enterprise resource
planning server of the buyer and a dilution prediction for the buyer‐supplier pair. Historical
dilution events from the enterprise resource planning system of the buyer can include automatic a
nalysis of the data of historic payment amounts for confirmed invoices and corresponding gross amounts
for the same confirmed invoices issued by the supplier over a predetermined time period. [00026] Additionally, in some systems and methods of the inv
ention, the dilution prediction for the buyer‐supplier pair is based on the use o
f a machine learning (ML) model trained by billings
and correspondent payments from 1st tier buyers to t
heir suppliers in different industries, different jurisdictions, and different economic cycles. Often,
the billings and payment amounts used for the machine learning (ML) model are significant (on the
order of $ trillions). The machine learning models can then be used to further process the fast
data, including billings of suppliers and payments
of buyers, to determine the dilution prediction. Th
e dynamic credit limit server can include machine learning (ML) modules for receiving the dilution pred
iction generated by the models and to incorporate the information in determining the dynamic
credit limit. [00027] Systems and methods in accordance with the invention
include dynamic credit limit servers and applications that determine the risk scor
e multiplier in a number of ways. For example, the risk score multiplier can be based on one or m
ore of a financial analysis score, a fraud‐threat
score, a compliance score, a business credit score,
a filed liens score, a success‐social value score,
and a track record score. [00028] In some example embodiments of the invention, a fina
ncial analysis score is based on one or more of a profit and loss statement, a bala
nce sheet, an accounts receivable aging report, and
an accounts payable aging report received from the s
upplier’s accounting software. Financial analysis scores can also be based on a number of o
ther factors, including an extrapolated income determination, an equity‐income ratio, an assets‐li
ability ratio, a monthly revenue ratio, a total accounts payable‐accounts receivable comparison, a ti
me‐based accounts payable‐accounts receivable comparison, a highest accounts receivable c
oncentration, and/or a reporting period. [00029] Example implementations of the invention can base a
fraud‐threat score on one or more of an email risk based on an email address ag
e, an IP address confidence based on a historical IP address fraudulent use, and an IP address risk b
ased on a location of an IP address and a supplier
location. Other bases for a fraud‐threat score ca
n include a Proxy‐VPN‐TOR determination based on
a direct or non‐direct connection, an email free‐
corporate determination based on whether a user is using corporate email or free email, and a city con
fidence score based on a location of a city and proximity to the supplier’s address. Similarly, a
fraud‐threat score can include considerations of a
geolocation score based on a supplier’s address and
IP address, a location accuracy radius score based on a supplier business location and user IP a
ddress distance, and a location average income score based on a weighted average of income per per
son for a postal code associated with the IP address. Likewise, additional considerations for a f
raud‐threat score can include a postal confidence score based on a business account address or a cred
it card address and a user address, and an address‐phone residential‐business score based on a
user connection location. [00030] Systems and methods in accordance with the invention
can determine a compliance score based on a number of factors, including one o
r more of an international watchlist score based on an Office of Foreign Assets Control sanctions lis
t, an enhanced credit score based on a combination of data from one or more registered cred
it agencies (and augmented with utility records, electoral rolls, and drivers’ license recor
ds), and a passport‐driver license‐ID validation s
core (digital identity intelligence). Further, a complianc
e score can also be based on an address validation
score, and a utility score based on correlation of
utility bills with a user address. [00031] Some implementations of the invention include dynami
c credit limit servers and application that determine a business credit score ba
sed one or more of an active registration‐time in
business score, a derogatoriness score, an insolvency
history score, a collection‐revenue ratio score, a
tax liens‐CCJ (county court judgment) history score,
and a trade names score, where these scores are based on credit reports in data‐feed format receive
d from a credit bureau. [00032] Example systems and methods in accordance with the i
nvention determine a filed liens score based on at least one of a number of
not‐terminated filings (UCC/PPSR/Charges) score (i.e., Uniform Commercial Code, Personal Property Secu
rities Register, etc.), a number of not‐ terminated filings (AR/Debtors) score, a not‐terminat
ed filings (Inventory) score, a not‐terminated filings (PMSI) score, a not‐terminated filings (All
Assets) score, any or all of which can be based o
n collateral descriptions. An alternate payee score ba
sed on existence of an active payment assignment to a third party can also be used in de
termining a filed liens score. [00033] In some embodiments of the invention, the dynamic cr
edit limit servers and applications can determine a success‐social value sc
ore based on one or more of a level of education
score, a career score, a loyalty score (based on fr
equency and duration of job changes), a Social Activity score (based on user social involvement in
charities, community activities, and/or military background), and a reputation score (based on collect
ed data related to user‐signer reputation). [00034] In some implementations of the invention a loyalty s
core can include a level of continued effort to achieve a goal and a measure of
jumping back and forth from project to project.
[00035] Systems and methods in accordance with the invention
can determine a track record score, where the track record score is with the sup
ply chain finance provider and is based on at least
one of a longevity of account score, a defaults num
ber to transactions number ratio, a ratio of post‐
confirmation dilution to gross amount of all confirme
d invoices, a ratio of post‐confirmation dilution to a total of an amount of automatic recourse avail
able in the current payment period, an amount of automatic recourse available in the next payment peri
od, and a total confirmed invoices score. In some implementations of the invention, a post‐confir
mation dilution can include at least one of the set of an amount of unrelated credit‐memos applied
on the same date as an invoice scheduled payment date, an amount of chargebacks, withholdings,
counterclaims and an amount of set‐offs. [00036] Any or all of the scores making up the risk score
multiplier can be statically or dynamically weighted, for example, based on received
fast data and other information received by the dynamic credit limit server. The scores making
up the risk score multiplier can be dynamically weighted using different multipliers applied to each
score to differentiate contributions of each score to the overall risk score multiplier. [00037] In addition, in some example systems and methods of
the invention, risk score multipliers are also based on a number of other fac
tors including one or more of an Authentication Result score, an ID Passed/Failed/Unknown score, a To
tal ID Verification score, an ID Barcode Verification score, and an ID Data Extraction Reliabi
lity Level score. These factors contributing to digital identity intelligence can be based on analysi
s of driver licenses or passports for a predetermined jurisdiction. BRIEF DESCRIPTION OF DRAWINGS [00038] The patent or application file contains at least one
drawing executed in color. Copies of this patent or patent application publication with
color drawing(s) will be provided by the Office upon request and payment of the necessary fee. [00039] FIG. 1 is a diagram of a system for creating a dy
namic credit limit and related automatic recourse base for providing supply chain fi
nance services in accordance with the invention. [00040] FIGS. 2A‐2B are flow diagrams of a method for cre
ating an initial dynamic credit limit and recourse base for providing supply chain finance
services in accordance with the invention as well as for subsequent supply chain finance funding.
[00041] FIG. 3 is a flow diagram of an example dynamic cre
dit limit determination in accordance with the invention and showing relationship
s between system components and scores. [00042] FIG. 4 is an example user interface screen in accor
dance with the invention showing supplier interactions with a system for creating a d
ynamic credit limit and related automatic recourse base. [00043] FIG. 5 is an example user interface screen showing
supplier interactions with a system for creating a dynamic credit limit and relat
ed automatic recourse base using a Cash Planner feature. [00044] FIG. 6 is an example user interface screen showing
supplier interactions with a system for creating a dynamic credit limit and relat
ed automatic recourse base using an Always‐Pay‐ Me‐Early feature. DETAILED DESCRIPTION [00045] The systems and methods of the invention provide tec
hnical solutions for processing and analyzing disparate data in large volumes at sca
le from multiple sources. In analyzing the multitude of financial data quickly, accurately, and
efficiently, dynamic credit limits and recourse bases are established in real time for suppliers.
Dynamic credit limits determined in accordance with the invention are responsive to changes in informatio
n, including financial information from the buyer and the seller along with many other types of
fast data received and evaluated by the dynamic credit limit systems and methods. The determinations
of the dynamic credit limits reliably allow banks and other financial institutions to offer suppl
y chain finance (SCF) service without requesting a payment guaranty from a buyer, which is traditionally
used as a hedge against the risks of non‐ payment and payment dilution. By forgoing a payment
guaranty from a buyer, supply chain finance services are more widely available and provide improv
ed liquidity and working capital without effecting changes to accounting treatment. [00046] The dynamic credit limit enables early payment servic
es and solutions against confirmed invoices. As the fast data to the supply
chain finance provider change, the dynamic credit limit for the supplier may also change. Similarly,
as the number and volume/amount of confirmed invoices changes, the dynamic credit limit available
to that supplier may also change. The invention described in this disclosure provides a technological
solution in improved systems and methods for receiving, managing, and evaluating extensive and dist
inct data sets from a wide array of sources. This invention includes systems and methods that auto
matically create a dynamic credit limit at the point‐of‐funding decision based on analysis of cri
tical risk factors identified in the fast data obtai
ned in real time. The systems and methods of the inve
ntion also automatically structure a recourse base, allowing supply chain finance providers to mitigate n
on‐payment and dilution risks. These systems and methods allow supply chain finance funders to re
liably and safely provide supply chain finance services without requiring a buyer payment guaranty.
System Overview [00047] Example embodiments of the invention feature systems
and methods for creating a dynamic credit limit. FIG. 1 shows a block diagram
of a multimodality supply chain finance provider system 100 for receiving requests for supply chain f
inance services, automatically receiving confirmed invoices, automatically receiving fast data,
and determining the dynamic credit limit. System 100 includes data exchange network 199. Data
exchange network 199 is the medium used to provide communications links between various devices a
nd computers connected together within the system 100. Data exchange network 199 can incl
ude connections, such as wire, wireless communication links, or fiber optic cables, from indi
vidual clients, servers, sources of fast data, and processing components. The clients, servers, data so
urces, and processing components can access the data exchange network 199 using different softwar
e architectural frameworks, different web services, different file transfer protocols, and diffe
rent Internet exchange points. Data exchange network 199 can represent a collection of networks a
nd gateways that use the Transmission Control Protocol/Internet Protocol (TCP/IP) and other communica
tion protocols, as well as application programming interfaces (APIs), to communicate with one
another and with devices connected to the Data exchange network 199. One example data exchang
e network 199 includes the Internet, which can include data communication links between major no
des and/or host computers, including thousands of commercial, governmental, educational, and
other computer systems that route data and messages. FIG. 1 is one example of an environ
ment of the invention and is not an architectural limitation for different illustrative embodiments of t
he invention. [00048] Clients and servers are only example roles of certai
n data processing systems and computer systems connected to data exchange network 1
99 which do not exclude other configurations or roles for these data processing sys
tems. Dynamic credit limit server 130 and buyer server 150 connect to data exchange network 199 alon
g with sources of fast data 160, 162, 164, 166, 168, 170 (which can include servers, databases, proce
ssors, and the necessary software and hardware to execute applications and methods for acqu
iring and sending fast data). Software applications can execute on any computer in the syst
em 100. User computers (clients), including supplier computing device 110, are also connected to
data exchange network 199. A data processing (computer) system, such as servers 130, 150 and clie
nts 110, and data sources 160, 162, 164, 166, 168, 170 (and other connected devices) can include d
ata and can have software applications and/or software tools executing on them. [00049] FIG. 1 shows an example system architecture and show
s certain components that are usable in an exemplary implementation of the inventio
n. For example, servers 130, 150, 160, 162, 164, 166, 168, 170 and client 110 are depicted as
servers and clients only as example and not to imply a limitation to a client‐server architecture.
In another example embodiment of the invention, the system 100 can be distributed across several dat
a processing (computer) systems and a data network as shown. Similarly, in another example emb
odiment of the invention, the system 100 can be implemented on a single data processing system wi
thin the scope of the illustrative embodiments. Data processing (computer) systems 110, 130, 150, 160
, 162, 164, 166, 168, 170 also represent example nodes in a cluster, partitions, and other co
nfigurations suitable for implementing an embodiment of the invention. [00050] The supplier computers (e.g., 110) can take the form
of a smartphone, a tablet computer, a laptop computer, a desktop computer, a w
earable computing device, or any other suitable computing device and computers 130, 150, 160
, 162, 164, 166, 168, 170 are typically servers, personal computers, and/or network computers. Softwar
e application programs described as executing in the system 100 in FIG. 1 can be confi
gured to execute in user computers in a similar manner. Data and information stored or produced in
another data processing system can be configured to be stored or produced in a similar ma
nner. [00051] Applications 111, 131, 151 implement an embodiment or
function of the invention as described in this document. For example, dynamic cr
edit limit application 131 receives a request from an application 111 on supplier computing device
110, including payment information such as currency, payment dates, selected invoices, and other
supplier information. Applications 111, 112 of the supplier implements an embodiment or a function
as described to operate in conjunction with applications 131, 132 on the dynamic credit limit se
rver 130. For example, application 111 provides the supplier payment information used by dynamic cred
it limit application 131 to process, classify, and provide actionable dynamic credit limit funds.
Similarly, buyer application 151 operates in conjunction with application 131 on the dynamic credi
t limit server 130 and provides invoices and ERP system records used by dynamic credit limit appl
ication 131 to process, classify, and provide actionable dynamic credit limit funds. [00052] Computers 110, 130, 150, 160, 162, 164, 166, 168, 1
70, and additional computers (e.g., clients and servers), may couple to data exch
ange network 199 using wired connections, wireless communication protocols, or other suitable da
ta connectivity. [00053] In the depicted example, dynamic credit limit server
130 may provide data, such as boot files, operating system images, and applications
to user computers (clients and servers) 110, 150. Client 110 may be clients to server 130 in
this example. Client 110 and servers 150, 160, 162,
164, 166, 168, 170, or some combination, may include
their own data, boot files, operating system images, and applications. System 100 may include ad
ditional servers, clients, and other devices that are not shown. For example, while countless point
of sources of fast data can be used to provide inputs to the dynamic credit limit server using sy
stems and methods constructed according to the principles and exemplary embodiments of the invention,
for clarity and brevity, six distinct sources of fast data are shown with a single supplier computing
device, a single buyer server, and a single dynamic credit limit server as shown in FIG. 1.
[00054] Among other uses, system 100 may be used for implem
enting a client‐server environment in accordance with exemplary embodiments o
f the invention. A client‐server environment enables software applications and data to
be distributed across a network such that an application functions by using the interactivity betwe
en a user computer and a server. System 100 may also employ a service‐oriented architecture,
where interoperable software components distributed across a network can be packaged together
as coherent applications. [00055] Together, the system 100 provides inputs for the dyn
amic credit limit application 131 to process, classify, and provide actionable dynamic
credit limit recommendations. More specifically, the supplier computer device 110 can use a supplier
portal to provide to provide payment requests for chosen invoices. The system 100 uses inputs fr
om buyer server 150, including ERP system 153, for example, as further inputs for the dynamic credi
t limit application 131. Additionally, sources 160,
162, 164, 166, 168, 170 provide fast data in real
time for use in determining the dynamic credit limi
t. As inputs and results from supplier computing device
110, buyer server 150, and sources of fast data 160, 162, 164, 166, 168, 170 change in real‐time,
dynamic credit limit server 130 and application 131
constantly and continually reassess and redetermine th
e actionable dynamic credit limit. Glossary [00056] The example systems and methods in accordance with t
he invention may be better understood by providing contextual meanings of some o
f the terms used in the examples of this disclosure as shown below. [00057] L d – Dynamic Credit Limit (“DCL”) – is t
he maximum amount, at a specific moment in time (current moment), of confirmed and scheduled for
payment invoices issued to an individual Buyer that can be discounted/paid early by the suppl
y chain finance provider; [00058] I i –Issued invoice – amount of invoice issued
by a Supplier to a Buyer (Account Debtor); [00059] I v – Verified invoice – amount of invoice issu
ed by a Supplier and the invoices’ deliverables have been verified by the Buyer. Invoice
s with this level of approval are usually accepted
for invoice finance funding such as factoring, invoic
e discounting, asset‐based lending (ABL), etc.; [00060] I c – Confirmed Invoice – amount of invoice a
pproved and scheduled for payment. Invoices with this level of approval are usually acc
epted for supply chain finance I c = I v – Δ v where Δ v is pre‐confirmation dilution (e.g., an amount
of chargebacks, set‐offs, counterclaims, withholdings, etc.); [00061] I e – early paid by supply chain finance confirme
d invoice I c [00062] G c – Gross amount of all confirmed invoices;
[00063] G e – gross amount of all early paid by supply
chain finance invoices; [00064] P c – Amount of payment received by supply chai
n finance provider for confirmed invoice P c = I c – Δ c where Δ c is the post‐confirmation dilution (e.g., amou
nt of unrelated credit‐memos applied on the same date as invoice scheduled payment date, chargeba
cks, set‐offs, etc.); [00065] P g – Amount of payment received by supply chai
n finance provider for all confirmed invoices G c ; [00066] D d – Invoice due date; [00067] D p ‐ invoice scheduled payment date; [00068] D e – supply chain finance early payment date;
[00069] T c – Current payment period; [00070] T n – next payment period; [00071] t d – decision moment (milliseconds); [00072] R c – Amount of automatic recourse available in
the current payment period T c R c = P g ‐ G e ; [00073] R n – Amount of automatic recourse available in
the next payment period T n . Projected available automatic recourse is an amount o
f payments to be received from the same Buyer for non‐funded invoices (P g above DCL L d ); [00074] C b – Base ceiling – initial maximum % of th
e total amount of G c available for funding by SCF and established for a specific Buyer by the
SCF underwriters; [00075] d a – Discount – amount of discount taken and
defined by supply chain finance provider and usually calculated as: d a = d % *I e *(D p – D e ) where d % is daily % discount; [00076] P e – Early payment amount of invoice paid by
supply chain finance provider on early payment date D e – discounted amount of I e P e = I e ‐ d a Example System Operation and Methods: [00077] The systems and methods of the invention establish a
Dynamic Credit Limit L d for a Supplier by performing real time (instant) and period
ic analysis of major non‐payment risks, including dilution risk, and create a dynamic risk mitigation
structure in the form of an automatic recourse capability, including R c the amount of automatic recourse available in
the current payment period and R n the amount of automatic recourse available in t
he next payment period. [00078] One example embodiment of the invention is shown in
the system diagram of FIG. 1 and the process flow diagram of FIG. 2. The dynam
ic credit limit determination begins after block 202 when a supplier signs up for supply chain finan
ce (SCF) service (A in FIG. 1). For clarity and
brevity, “supplier” and “supplier computing devic
e” may be thought of as a party selling goods or
services to a buyer, and an example computer system
used by a supplier is the supplier computing device 110 shown in FIG. 1. The process continues
in block 204 when supplier 110 requests supply chain finance (SCF) service from the dynamic credit
limit server 130 of supply chain finance provider (B in FIG. 1) [00079] In block 206, the dynamic credit limit server 130 o
f supply chain finance provider contacts the specific buyer (server) 150 involved in
the transactions with the supplier 110 (C1 in FIG.
1). The dynamic credit limit server 130 of supply
chain finance provider receives/pulls confirmed invoices from the specific buyer (server) 150 (e.g.,
from the buyer ERP system 153) and receives the total (gross) amount of all outstanding confirmed inv
oices G c (C2 in FIG. 1). [00080] The example process continues in block 208 as the d
ynamic credit limit server 130 of supply chain finance provider receives/pulls fast data
from a variety of sources of fast data 160, 162,
164, 166, 168, 170 (D1‐D6 in FIG. 1). As outlin
ed further below, in block 210, the dynamic credit
limit server 130 of supply chain finance provider determine
s the supplier’s Dynamic Credit Limit for the particular buyer 150 (E in FIG. 1) based on inputs
from the buyer server ERP system 153, accounting software 113 of the supplier, and fast data from so
urces 160, 162, 164, 166, 168, 170. In many example implementations of the invention, the accounti
ng software 113 of the supplier resides apart from the physical location of the supplier, as shown
in FIG. 1 and behaves similarly to the fast data
from sources 160, 162, 164, 166, 168, 170. FIG. 4
shows an example user interface screen 400 with the determined dynamic credit limit 411 [00081] Several contributing elements based on fast data anal
yzed by the dynamic credit limit application 131 can deliver binary conclusions
like “yes” or “no.” For example, if during
supply chain finance digital compliance processes (KYC, AML,
CTF—know your customer, Anti‐Money Laundering, and Counter Terrorism Financing, respective
ly) someone from a Supplier’s management or ownership team is listed on an Office of Foreign
Assets Control (OFAC) list, it will result in the
Supplier’s funding request being declined. On the o
ther hand, for example, the level of confidence in
a logged‐in Supplier user’s IP address or a user
’s geo‐location or analysis of UCC/PPSR/Charges (i
.e., Uniform Commercial Code, Personal Property Securities
Register, etc.) and collateral descriptions for a Supplier will contribute to the relevant segment’
s element (risk score) scoring, resulting in a decrease of the final current Dynamic Credit Limit L
d level, instead of rejection of the funding re
quest. [00082] In block 212, Supplier 110 requests payment for chos
en invoices G e (i.e., the gross amount of all early paid by supply chain finance in
voices) minus a discount. As an example, the invoices chosen are shown as reference numerals 413,
415, 417, 419 on the user interface screen 400 of FIG. 4, and the discount is shown as reference
numeral 441. The payment amount must be less than or equal to the dynamic Credit Limit 411 deter
mined by the dynamic credit limit server 130 of supply chain finance provider. The user interface s
creen 400 in FIG. 4 shows the invoice amounts selected 431 as the supplier identifies invoices for
early payment. Along with the payment amount, the supplier 110 choses an early payment date (F in
FIG. 1) and shown as reference numeral 451 in FIG. 4. [00083] The process continues in block 214 as the dynamic c
redit limit server 130 of supply chain finance provider provides (early) payment for t
he chosen invoices to the supplier 110 (G in FIG.
1). At the same time, in block 216, the dynamic
credit limit server 130 of supply chain finance provider determines an automatic recourse base R c , which is the amount of automatic recourse available in the current payment period, for the sup
plier as a difference between the gross amount of all outstanding confirmed invoices (G c ) and the gross amount of all early paid by
supply chain finance invoices G e (H in FIG. 1). [00084] In block 218, the dynamic credit limit server 130 o
f supply chain finance provider receives payment P g (i.e., the amount of payment received by the
supply chain finance provider for all confirmed invoices) from the buyer 150 for total
confirmed invoices G c and payment dilution Δ c calculated for all funded (confirmed outstanding) invo
ices G c (J in FIG. 1). [00085] In one example implementation of the invention, for
dilution risk calculations, the process performed by the dynamic credit limit server
130 of supply chain finance provider constantly and continually receives and analyzes historical dilut
ion data of the Supplier with the individual Buyer
and can use ML (machine learning) for dilution predi
ction providing systems the ability to automatically learn and improve from experience withou
t being explicitly programmed. Machine learning determinations of predicting dilution focuses
on the development of computer programs that can access data and use it learn to refine di
lution calculations and predictions themselves. [00086] For example, the dynamic credit limit server 130 may
receive fast data dilution predictions based on machine learning modules, applica
tions, and systems for identifying relationships and invoice/payment histories for use in
calculations by the dynamic credit limit server 130 for decisions relating to setting a dynamic cred
it limit. The dynamic credit limit server can includ
e machine learning (ML) modules for receiving the dilut
ion prediction generated by the models and to incorporate the information in determining the dynamic
credit limit. [00087] Beside the dilution risk, the systems and methods of
the invention manage a number of other types of Supplier risks, including but not
limited to, fraud/threat, credit risk, liens (includi
ng registered secured interests and collateral description
s), CCJs and pending litigations, tax liens, compliance, etc. [00088] These non‐dilution risks are important because they
can significantly impact current and future payments from the Buyer to the Supplier
and the ability to structure a sustainable recourse base to mitigate risks and prevent real los
ses. For example, understanding how stable the Supplier’s business is at the decision point t d and the ability to predict if the Supplier w
ill be able to deliver its goods and/or services to the Buyer next
month and beyond figures prominently in determining the dynamic credit limit. Examples of t
he consideration and effects of these risks are outlined further below. [00089] The overall process continues in FIG. 2B as addition
al comparisons, calculations, and determinations are made by the dynamic credit limit
server 130 of supply chain finance provider when supplier 110 is in need of additional funds.
For example, continuing with block 222 in FIG. 2B,
the dynamic credit limit server 130 of supply chain
finance provider compares the amount of automatic recourse available in the current payment p
eriod R c to Δ c the post‐confirmation dilution. If the post‐confirmation dilution is less than the aut
omatic recourse available in the current payment period, the process continues to block 224 as descri
bed below. However, if the post‐confirmation dilution is greater than the automatic recourse avail
able in the current payment period, the process continues to block 254, and the dynamic credit limit
is set to zero. The process continues to block
258 where the buyer’s payments for the next group
of confirmed invoices R n is used to cover the gross amount of all early paid by SCF invoices G e until dilution and fees are fully covered. Th
e process then continues to block 224. [00090] The process in block 224 determines if the supplier
requests additional supply chain finance service. If no, the supply chain finance s
ervice is suspended in block 262, and the process ends. However, when the supplier does need addition
al supply chain finance service in block 224, the process continues to block 228, where the supply
chain finance provider pulls all confirmed invoices from the buyer and total amounts of all ou
tstanding confirmed invoices G c . All data are pulled from all sources of fast data, and the suppl
ier’s dynamic credit limit is (re)calculated in blo
ck 232. [00091] In block 236, the supplier selects an early payment
date and requests (and receives) payment for (newly) chosen invoices G e less the discount, and the total is less tha
n the recalculated dynamic credit limit. In block 240, a new automati
c recourse Rc is created as a difference between the total amounts of all outstanding confirmed invoic
es G c and the gross amount of all early paid by SCF invoices G e . [00092] In block 244, the supply chain finance provider rece
ives payment P g for the total of all confirmed invoices G c and payment dilution Δ c that was calculated for all funded confirmed
invoices G c . The process then returns to block 222 and
cycles through until the supplier no longer needs additional supply chain finance service, and service
is suspended. [00093] An additional example with simplified amounts provides
additional insights into the systems and methods of the invention. One example
includes a Supplier’s request for early payment. At the specific moment t d , this Supplier has $1M of confirmed invoices
G c issued to a specific Buyer. As a result of the automatic analysis based on curr
ent fast data, including financial data pulled from
the Supplier’s accounting software, historical diluti
on data pulled from the Buyer’s ERP server, and data pulled from the different fast data sources, in
cluding Credit Bureaus, different government agencies, compliance/fraud/threat data aggregators, PPSR
/PPSA systems and UCC filing data pulled from Secretary of State databases, etc. The automat
ic analysis includes predicted dilution Δ c , where predicted dilution is calculated based on machine lea
rning (ML) modules trained by significant data of billing and correspondent payments ($ trillions) f
rom 1st tier Buyers to their Suppliers in different
industries, jurisdictions, economic cycles, etc. The
machine learning (ML) models can then be used to further process the fast data, including billings of
suppliers and payments of buyers, to determine the
dilution prediction. Example machine learning engines
(ML) include Google TensorFlow or BERT, and other engines. In this example, the dynamic credit
limit (DCL) L d is set at a level of $912K. [00094] This means that the Supplier can discount any invoic
es up to $912K and the supply chain finance provider will provide early payment P e to the Supplier on the early payment date D e and will take a discount payment of d a . For example, d a is calculated as 1% for 30 days daily prorat
ed discount. All payments from the Buyer for all invo
ices, P g are assigned and paid to a dedicated account controlled by the supply chain finance provid
er (via assignment of debts). So, in an event where payments for the funded $912K invoices are dil
uted, the dilution Δ c will be covered automatically from payments received for $88K in non
funded invoices R c . If the amount of current recourse R c is not enough to cover the dilution Δ c, then the dynamic credit limit (DCL) will be se
t to $0 and deficit will be covered from the next payment p
eriod for non‐funded invoices R n as shown in FIG. 2B. Immediately after this, the Supplier will again
be able to request early payments under the new dynamic credit limit DCL L d . [00095] The systems and methods of the invention provide dig
ital supply chain finance service on a recourse basis, and in cases where the
supply chain finance provider is unable to cover dilution Δ c by automated current recourse R c and next payment period recourse R n , they can exercise full recourse rights and demand payments from any ot
her receivables that the Supplier has or will have in future. The supply chain finance provider
registers its security interest and is a secure creditor to the Supplier. Dynamic Credit Limit Determination Example: [00096] FIG. 3 shows an example of a scoring scheme to qua
ntify aspects of fast data received by the dynamic credit limit and recourse ba
se systems in accordance with the invention. In the example shown in FIG. 3, the dynamic credit lim
it 311 is determined based on a total amount of all outstanding confirmed invoices 328, a historic di
lution multiplier 322, a buyer base ceiling 324, and a risk score multiplier 326. FIG. 3 can be u
nderstood as showing a loose hierarchy of scores/sub
scores and sources of fast data that contribute to
the dynamic credit limit determinations. Successive lower “layers” in FIG. 3 show elements
contributing to the components above them. As outlined above, a dynamic credit limit L d for a specific Supplier is calculated at ever
y decision moment t d and is based on real time fast data pulled
from thousands of data points, for example financial
information from Supplier accounting software, data fr
om Credit Bureaus, and other sources of fast data as shown in FIGS. 1 and 3. In one example
implementation of the invention, a dynamic credit limit L d 311 is determined using a gross amount of al
l confirmed invoices G c 328 as well as a historic dilutions multiplier M h 322 (see below), a risk score multiplier M s 326 (also described further below), and a base ceiling C b 324. Both the historic dilutions multiplier M h 322 and the risk score multiplier M s 326 are applied to the base ceiling C b 324. The base ceiling C b 324 is an initial maximum % of the total gross amount of all confirmed invoices availabl
e for funding by the supply chain finance system.
The base ceiling C b 324 is established for a specific Buyer by t
he supply chain finance underwriters 334. In an example embodiment of the invention, th
e dynamic credit limit 311 is determined based on the following relationship. L d = G c *(C b *M h *M s ) where M h – historical dilutions multiplier (in %) whi
ch is applied to base ceiling C b . [00097] Continuing with FIG. 3, M h 322 is calculated as the result of statistica
l analysis of historical dilution events 332 for a period, based o
n data constantly pulled from the Buyer ERP system 392, and contributions from ML dilution predic
tion for specific Buyer‐Supplier pairs as reference numeral 393 shown in FIG. 3. The dilution
prediction 333 is determined based on machine learning (ML) trained by significant data of billing
and correspondent payments (in $ trillions) from 1st tier Buyers to their Suppliers in different indu
stries, jurisdictions, economic cycles, etc., The machine learning models can then be used to further
process the fast data, including billings of suppliers and payments of buyers, to determine the d
ilution prediction 333. Example machine learning models engines (ML) include Google TensorFlow
, Google BERT, and other engines. M s – risk score multiplier (in %) which is ap
plied to base ceiling C b . [00098] Continuing with the example dynamic credit limit dete
rmination outlined in the FIGS., risk score multiplier M s 326 is equal to a percent from 0% to 100%
depending on what the total risk score 336 is. For example, if the total risk
score S t 336 is higher than 450, then the risk score
multiplier M s 326 is equal to 100%. Similarly, if the to
tal risk score S t score 336 is lower than 25, then the risk score multiplier M s 326 is equal to 0%. The supply chain finan
ce provider can establish the strata of risk score multipliers M s 326 equating them to the total risk scores S
t 336. [00099] As outlined above, total risk score S t 336 is used to determine the risk score multiplier 326. In turn, the total risk score S t 336 is made up of many different scores dete
rmined from the constant and continuing analysis of fast da
ta. In one example implementation of the invention, the total risk score 336 is determined ba
sed on a financial analysis score 342, a fraud/threat (trust) score 343, a compliance score 34
4, a business credit score 345, a filed liens score
346, a success value score 347, an internal track r
ecord score 348, and other scores based on received fast data. [000100] While the financial analysis score 342 and the succe
ss value score 347 are shown in FIG. 3 as optional, the total risk score S t 336 can include more or less fast data from
many sources. [000101] In an example embodiment of the invention, the total
risk score 336 is determined based on the following relationship. S t – is the total risk score (maximum value i
s 500) S t = (S f *W 1 + S t *W 2 + S g *W 3 + S c *W 4 + S l *W 5 + S s *W 6 + S i *W 7 + … S x *W x )/Σ(W i ) where S f is Financial Analysis score, S t is Fraud/Threat (trust) score, S g is Compliance score, S c is Business Credit score, S l is Filed Liens score, S s is Success/Social Value score, S i is SCF Internal Historical Track Record score,
S x is any additional customized score(s) which co
uld be added, if necessary, by the supply chain finance provider and the weighting facto
rs (W 1 , W 2 , etc.) are applied to each of the respective scores. W i is a weight factor, which has a value from
0 to 5 and could be dynamic, depending on the value of the applicable score. [000102] Each of the scores can be weighted by the supply c
hain finance provider and assigned a weight factor. For example, weighting the individ
ual scores (e.g., financial analysis score, fraud/threat (trust) score, compliance score, business
credit score, filed liens score, success value score, internal track record score, and customized sc
ores) that make up the total risk score provides a
method of prioritizing scores and determining a relat
ive value of that score’s contribution to the tota
l risk score. In one example implementation of the i
nvention, criteria for weighting is selected by analyzing statistics and identifying characteristics of
each score’s correlation to the total risk score.
The weight for each score is assigned in the range
from 0 to 5. The most important score (i.e., tha
t score or scores that have the strongest correlation
to the total risk score and make the largest contribution to the total (aggregate) risk score) is
weighted 5, and less important scores (i.e., that score or scores that have the weakest correlation to
the total risk score and makes the smallest contribution to the total aggregate score) is weighte
d 1. The weights can be statically applied to each
of the scores, sub‐scores, sub‐sub‐scores, etc.
or can be dynamically determined and applied based on received fast data. [000103] The individual scores 342, 343, 344, 345, 346, 347,
348 are based on the received fast data that is continually processed by the system.
In some instances, the system determines a number of sub‐scores for the scores based on the
fast data. In FIG. 3, an example is shown where
Financial Analysis Score 342 is determined based on
sub‐scores (and similarly, on sub‐sub‐scores).
While determining a Financial Analysis score S f is optional, it can be calculated based on a
nalysis of the Supplier’s financial data, if available, includi
ng Profit and Loss Statements, Balance Sheets, AR Aging and AP Aging reports, and other financial info
rmation instantly pulled from the Supplier’s accounting software (e.g., 113 in FIG. 1) or automat
ically extracted from Suppliers reports uploaded to the supply chain finance provider’s platform in
any format, including from sources of fast data.
In FIG. 3, a Profit and Loss Statement (shown as refer
ence numeral 362 in FIG. 3) may result in a sub‐
sub‐score 352. Similarly, fast data (shown as ref
erence numeral 363) may be used to determine another sub‐sub‐score 353, such as an A/R concent
ration, or other measures of the supplier’s financial position. [000104] In practice, the financial analysis score S f 342 may be more important for invoice finance services than for supply chain finance servic
es, since there is no verifiable historical dilution
data which exists for invoice finance. That is why
for supply chain finance services, financial analysis
score S f 342 may not be as important as buyer histori
cal dilution analysis 332 (and the resulting historical dilution multiplier 322), which is a more
determinative component for the dynamic credit limit determination. In any case, when a financial a
nalysis score S f 342 is available for supply chain finance services then it is calculated as: S f = (S f1 *W 1 + S f2 *W 2 + S f3 *W 3 + S f4 *W 4 + S f5 *W 5 + S f6 *W 6 + S f7 *W 7 + S f8 *W 8 + S f9 *W 9 + S f10 *W 10 + S f11 *W 11 )/ Σ(W i ) where S f1 is extrapolated income score, that can be bas
ed on past earnings statements, income from previous time frames, and other measures.
S f2 is Equity/Income ratio score, S f3 is Assets/Liability ratio score, S f4 is AR/Monthly Revenue ratio score, S f5 is Total AP/AR ratio score, S f6 is AP/AR 0 to 30 days ratio score, S f7 is AP/AR 30 to 60 days ratio score, S f8 is AP/AR 60 to 90 days ratio score, S f9 is AP/AR over 90 days ratio score, S f10 is Highest AR Concentration score, S f11 is Reporting Period score and W i is a weight factor. [000105] Each of the sub‐scores can be weighted by the sup
ply chain finance provider, and the weighting factors (W 1 , W 2 , etc.) are applied to each of the respective
sub‐scores. As before, weighting the individual sub‐scores that make up th
e total financial analysis score provides a method of prioritizing sub‐scores and determining a relativ
e value of that sub‐score’s contribution to the
total financial analysis score. In one example implementat
ion of the invention, criteria for weighting is selected by analyzing statistics and identifying chara
cteristics of each sub‐score’s correlation to the
total financial analysis score. The weight for ea
ch sub‐score is assigned in the range from 1 to
5. The most important sub‐score (i.e., that sub‐score or
sub‐scores that have the strongest correlation to
the total financial analysis score and make the larg
est contribution to the financial analysis score) is
weighted 5, and less important scores (i.e., that su
b‐score or sub‐scores that have the weakest correlation to the total financial analysis score and
makes the smallest contribution to the financial analysis aggregate score) is weighted 1. The weights
can be statically applied to each of the sub‐ scores, or can be dynamically determined and applied
based on received fast data. [000106] The fraud/threat score S t 343 is used in the dynamic credit limit dete
rmination to assess, identify, understand, and ultimately account f
or risks of fraud to the supply chain finance service provider. These risks can include fraudulent
disbursements, undisclosed relationships, theft by cyber‐fraud, false qualifications or certification
s, compliance with government regulations, improper reporting and disclosures, identity theft, an
d other fraud risks. The fraud threat score S t 343 is based on received fast data that is continua
lly processed by the system. In some instances, th
e system determines of sub‐scores that, in aggregate,
make up the fraud/threat score. In FIG. 3, an example is shown where fraud threat score S t 343 is determined based on sub‐scores (e.g.,
the sub‐ scores s t1 ‐s t11 described below). In FIG. 3, the arrows ext
ending downward from fraud/threat score 343 indicate further sub‐sub‐scores and respective
sources of fast data used to determine the sub‐ sub‐scores). In one example of the invention, det
ermining a Fraud/Threat Score S t is calculated based on the following relationship: S t = (S t1 *W 1 + S t2 *W 2 + S t3 *W 3 + S t4 *W 4 + S t5 *W 5 + S t6 *W 6 + S t7 *W 7 + S t8 *W 8 + S t9 *W 9 + S t10 *W 10 + S t11 *W 11 )/ Σ(W i ) where S t1 is Email Risk score, calculated based on the
email address age, S t2 is IP Address Confidence score, calculated bas
ed on the data for historical IP address fraudulent use, S t3 is IP Address Risk score, calculated based on
location IP address and Supplier location, S t4 is Proxy/VPN/TOR score, calculated based on an
alysis of direct or non‐direct connection, S t5 is Email Free/Corporate score, calculated based
on a fact that user is using corporate email address or free email address like g‐mail, yahoo,
etc., S t6 is City Confidence Score, calculated based on
location of city and proximity to Supplier’s address, S t7 is Geolocation score, calculated based on phys
ical user’s address and IP address, S t8 is Location Accuracy Radius score, calculated
based on Supplier’s business location and user IP address distance (radius), S t9 is Location Average Income score, calculated b
ased on the weighted average income in US dollars per person for the zip/post code(s) associate
d with the IP address, S t10 is Postal Confidence score, calculated based o
n business accounts (credit cards) address and user address, S t11 is Address/Phone Residential/Business score, cal
culated based on user connection location (office or home), and W i is a weight factor. [000107] Each of the sub‐scores can be weighted by the sup
ply chain finance provider, and the weighting factors (W 1 , W 2 , etc.) are applied to each of the respective
sub‐scores. As before, weighting the individual sub‐scores that make up th
e total fraud/threat score provides a method of prioritizing sub‐scores and determining a relative v
alue of that sub‐score’s contribution to the tota
l fraud/threat score. In one example implementation of
the invention, criteria for weighting is selected by analyzing statistics and identifying chara
cteristics of each sub‐score’s correlation to the
total fraud/threat score. The weight for each sub
‐score is assigned in the range from 1 to 5. T
he most important sub‐score (i.e., that sub‐score or
sub‐scores that have the strongest correlation to
the total fraud/threat score and make the largest co
ntribution to the fraud/threat score) is weighted 5, and less important scores (i.e., that sub‐score
or sub‐scores that have the weakest correlation to
the total fraud/threat score and makes the smallest
contribution to the fraud/threat aggregate score) is weighted 1. The weights can be statically applied
to each of the sub‐scores, or can be dynamically
determined and applied based on received fast data.
[000108] The compliance score S g 344 is used in the dynamic credit limit dete
rmination to assess, identify, understand, and ultimately account f
or risks of non‐compliance. These risks can include know your customer (“KYC”) non‐compliance
, anti‐money laundering (“AML”) non‐ compliance, counter‐terrorism financing (“CTF”) no
n‐compliance and other compliance risks. The compliance score S g 344 is based on received fast data that is
continually processed by the system. In some instances, the system determines of sub‐scores
that, in aggregate, make up the compliance score S g 344. In FIG. 3, an example is shown where
compliance score S g 344 is determined based on sub‐scores (e.g., the sub‐scores s g1 ‐s g5 described below). In FIG. 3, the arrows ext
ending downward from compliance score 344 represent further sub‐sub
scores and respective sources of fast data used to determine the sub‐sub‐scores). In one example
of the invention, determining a Compliance Score S g is determined by the following relationship: S g = (S g1 *W 1 + S g2 *W 2 + S g3 *W 3 + S g4 *W 4 + S g5 *W 5 )/ Σ(W i ) where S g1 is International Watchlist score, calculated ba
sed on OFAC sanctions list and/or international sanctions lists, S g2 is Enhanced Credit score, calculated based on
combination of the data from one or more registered credit agencies and is augmented with
one or more of the following; utility, electoral roll, and driver’s license or passport’s
data, S g3 is Passport/Driver License/ID Validation score,
calculated based on validation of personal IDs, S g4 is Address Validation score, calculated based
on user home address validation, S g5 is Utility score and calculated based on corr
elation of Utility bills with user address, W i is a weight factor. [000109] Similar to the above determinations, each of the sub
‐scores can be weighted by the supply chain finance provider, and the weighting fact
ors (W 1 , W 2 , etc.) are applied to each of the respective sub‐scores. As before, weighting the in
dividual sub‐scores that make up the total compliance score provides a method of prioritizing su
b‐scores and determining a relative value of that sub‐score’s contribution to the total complia
nce score. In one example implementation of the invention, criteria for weighting is selected by anal
yzing statistics and identifying characteristics of each sub‐score’s correlation to the total complian
ce score. The weight for each sub‐score is ass
igned in the range from 1 to 5. The most important sub
‐score (i.e., that sub‐score or sub‐scores that
have the strongest correlation to the total compliance sco
re and make the largest contribution to the compliance score) is weighted 5, and less important
scores (i.e., that sub‐score or sub‐scores that have the weakest correlation to the total compliance
score and makes the smallest contribution to the compliance aggregate score) is weighted 1. The w
eights can be statically applied to each of the sub‐scores, or can be dynamically determined and ap
plied based on received fast data. [000110] While outlined above, each sub‐component of the con
tributing component score has its own contributing sub‐sub‐components, and the s
ub‐sub‐scores corresponding to that sub‐sub‐ component combine to determine the aggregate sub‐sco
re. As alluded above, for example, Passport/Driver License/ID Validation score S g3 can be determined according to the relationshi
p: S g3 = (S g31 *W 1 + S g32 *W 2 + S g33 *W 3 + S g34 *W 4 + S g35 *W 5 )/ Σ(W i ) where S g31 is Authentication Result score, S g32 is ID Passed/Failed/Unknown score, S g33 is Total ID Verification score, S g34 is ID Barcode Verification score, S g35 is ID Data Extraction Reliability Level score
and W i is a weight factor. [000111] As with the determinations of the sub‐scores and t
heir corresponding weights, each of the sub‐sub‐scores can be weighted by the sup
ply chain finance provider, and the weighting factors (W 1 , W 2 , etc.) are applied to each of the respective
sub‐sub‐scores. As before, weighting the individual sub‐sub‐scores that make up the total
Passport/Driver License/ID Validation sub‐score S g3 provides a method of prioritizing sub‐sub‐scores a
nd determining a relative value of that sub‐sub‐ score’s contribution to the total Passport/Driver Li
cense/ID Validation sub‐score S g3 . In one example implementation of the invention, criteria for weightin
g is selected by analyzing statistics and identifying characteristics of each sub‐sub‐score’
s correlation to the total Passport/Driver License/ID
Validation sub‐score S g3 . The weight for each sub‐sub‐score is
assigned in the range from 1 to 5. The most important sub‐sub‐score (i.e., that sub‐sub
score or sub‐sub‐scores that have the strongest
correlation to the total Passport/Driver License/ID Va
lidation sub‐score S g3 and make the largest contribution to the Passport/Driver License/ID Validati
on sub‐score S g3 ) is weighted 5, and less important scores (i.e., that sub‐sub‐score or sub
sub‐scores that have the weakest correlation to t
he total Passport/Driver License/ID Validation sub‐score
S g3 and makes the smallest contribution to the Passport/Driver License/ID Validation aggregate sub‐sc
ore S g3 ) is weighted 1. The weights can be statically applied to each of the sub‐scores, or c
an be dynamically determined and applied based on received fast data. [000112] The (business) credit risk score S c 345 is used in the dynamic credit limit determination to quantify and account for risks of a
supplier’s overall ability to repay in case of recourse. To assess credit risk, the systems and m
ethods of the invention evaluate fast data related to credit history, ability to continue to provide se
rvice to Buyer, capacity to repay. The credit risk
score S c 345 is based on received fast data that is
continually processed by the system. In some instances, the system determines of sub‐scores that,
in aggregate, make up the credit risk score S c 345. In FIG. 3, an example is shown where credit
risk score S c 345 is determined based on sub‐scores (e.g., the sub‐scores s c1 ‐s c6 described below). In FIG. 3, the arrows ext
ending downward from credit risk score S c 345 represent further sub‐sub‐scores and re
spective sources of fast data used to determine the sub‐sub‐scores). In one example of
the invention, determining a Credit Risk Score S c 345 is determined by the following relationship: S c = (S c1 *W 1 + S c2 *W 2 + S c3 *W 3 + S c4 *W 4 + S c5 *W 5 + S c6 *W 6 )/ Σ(W i ) where S c1 Active Registration/Time in Business score, cal
culated based on time in business, S c2 is Derogatoriness score, calculated based on a
nalysis of derogatoriness data, S c3 is Insolvency History score, calculated based
on historical insolvency events, S c4 is Collection/Revenue Ratio score, calculated b
ased on the current amount in collection to Revenue, S c5 is Tax Liens/CCJ History score, calculated bas
ed on analysis of the liens placed against the supplier and historical judgments against the supplier
, S c6 is Trade Names Score, calculated based on the
history and numbers of the trade names used by supplier, and W i is a weight factor. [000113] Similar to the above determinations, each of the sub
‐scores can be weighted by the supply chain finance provider, and the weighting fact
ors (W 1 , W 2 , etc.) are applied to each of the respective sub‐scores. As before, weighting the in
dividual sub‐scores that make up the total credit
risk score provides a method of prioritizing sub‐sc
ores and determining a relative value of that sub‐
score’s contribution to the total credit risk score
. In one example implementation of the invention,
criteria for weighting is selected by analyzing stati
stics and identifying characteristics of each sub‐ score’s correlation to the total credit risk score.
The weight for each sub‐score is assigned in
the range from 1 to 5. The most important sub‐score
(i.e., that sub‐score or sub‐scores that have th
e strongest correlation to the total credit risk score
and make the largest contribution to the credit ris
k score) is weighted 5, and less important scores (i.e
., that sub‐score or sub‐scores that have the weakest correlation to the total credit risk score a
nd makes the smallest contribution to the credit risk aggregate score) is weighted 1. The weights can
be statically applied to each of the sub‐scores,
or can be dynamically determined and applied based on r
eceived fast data. [000114] The filed liens score S l 346 is used in the dynamic credit limit dete
rmination to quantify and account for risks of a supplier’s ass
ets being subject to another party’s security inter
est and their priorities. Filed liens imply a claim or
legal right against assets that are typically used
as collateral to satisfy a debt. If the underlying ob
ligation is not satisfied, the creditor may be able
to seize the assets that are the subject of the lien.
To assess filed liens and pledged collaterals, the
systems and methods of the invention evaluate fast d
ata related to lien filings, and the various types
of collateral identified in the lien. The filed li
ens score S l 346 is based on received fast data that is
continually processed by the system. In some instan
ces, the system determines of sub‐scores that, in
aggregate, make up the filed liens score S l 346. In FIG. 3, an example is shown where
filed liens score S l 346 is determined based on sub‐scores (e.g.,
the sub‐scores s l1 ‐s l6 described below). In FIG. 3, the arrows extending downward from filed liens score S l 346 represent further sub‐sub‐scores and respective sources of fast data used to determine th
e sub‐sub‐scores). In one example of the invention, determining a Filed Liens Score S l is calculated as according to the following r
elationship: S l = (S l1 *W 1 + S l2 *W 2 + S l3 *W 3 + S l4 *W 4 + S l5 *W 5 + S l6 *W 6 )/ Σ(W i ) where S l1 Total Number of Not‐Terminated Filings (UCC/P
PSR/Charges) score, calculated based on filed number of security interest liens, S l2 is Number of Not‐Terminated Filings (AR/Debto
rs) score, calculated based on filed number of security liens with collateral description which i
ncludes Accounts or Accounts Receivable or Debtors, S l3 is Not‐Terminated Filings (Inventory) score,
calculated based on filed number of security liens with collateral description which includes inven
tory, S l4 is Not‐Terminated Filings (PMSI) score, calcu
lated based on filed number of security liens with collateral description which includes purchase mo
ney security interest (“PMSI”), S l5 is Not‐Terminated Filings (All Assets) score,
calculated based on filed number of security liens with collateral description which includes all
assets, S l6 is the Alternate Payee Score calculated based
on existence of an active assignment of payment to third party, and W i is a weight factor. [000115] Similar to the above determinations, each of the sub
‐scores can be weighted by the supply chain finance provider, and the weighting fact
ors (W 1 , W 2 , etc.) are applied to each of the respective sub‐scores to determine the overall filed
liens score S l 346. As before, weighting the individual sub‐scores that make up the total filed
liens score S l 346 provides a method of prioritizing sub‐scores and determining a relative value of that
sub‐score’s contribution to the total filed lien
s score S l 346. In one example implementation of the i
nvention, criteria for weighting is selected by analyzing statistics and identifying characteristics of
each sub‐score’s correlation to the total filed
liens score S l 346. The weight for each sub‐score is a
ssigned in the range from 1 to 5. The most important sub‐score (i.e., that sub‐score or sub
scores that have the strongest correlation to the t
otal filed liens score S l 346 and make the largest contribution to the
filed liens score S l 346) is weighted 5, and less important scores (i.e., that sub‐score or
sub‐scores that have the weakest correlation to th
e total filed liens score S l 346 and makes the smallest contribution to th
e filed liens aggregate score S l 346) is weighted 1. The weights can be statically a
pplied to each of the sub‐scores, or can be dynamically determined and applied based on received
fast data. [000116] The success/social value score S s 347 is also used in the dynamic credit limit
determination to quantify and account for positive an
d negative effects which major shareholders and/or executive officers make on the supplier. The
goal of this metrics is to use positive social impact to increase inclusivity and access to capital
for suppliers. The success/social value score S s 347 is a proxy to an often qualitative value of the su
pplier, one that has been often measured with non‐
financial indicators. The success/social value sc
ore S s 347 is based on received fast data that is
continually processed by the system. In some instan
ces, the system determines of sub‐scores that, in
aggregate, make up the success/social value score S s 347. In FIG. 3, an example is shown where
success/social value score S s 347 is determined based on sub‐scores (e.g.,
the sub‐scores s s1 ‐s s56 described below). In FIG. 3, the arrows extending
downward from success/social value score S s 347 represent further sub‐sub‐scores and respective sou
rces of fast data used to determine the sub‐sub‐
scores). In one example of the invention, determini
ng a Success/Social Value score S s is calculated as according to the following relationship: S s = (S s1 *W 1 + S s2 *W 2 + S s3 *W 3 + S s4 *W 4 + S s5 *W 5 )/ Σ(W i ) where S s1 is Level of Education score, calculated based
on user historical education data, S s2 is Career score, calculated based on historica
l career data, S s3 is Loyalty score, calculated based on analysis
of professional commitment (level of continues effort to achieve the goal vs constantly j
umping from project to project), S s4 is Social Activity score, calculated based on
analysis of data connected to community, charities, and other social activities, S s5 is Reputation score calculated based on analys
is of user reputational data, and W i is a weight factor. [000117] As above, each of the sub‐scores can be weighted
by the supply chain finance provider, and the weighting factors (W 1 , W 2 , etc.) are applied to each of the respective
sub‐scores to determine the overall success/social value score S s 347. As before, weighting the individual sub
‐ scores that make up the total success/social value s
core S s 347 provides a method of prioritizing sub‐ scores and determining a relative value of that sub
score’s contribution to the total success/social value score S s 347. In one example implementation of the i
nvention, criteria for weighting is selected by analyzing statistics and identifying characteristics
of each sub‐score’s correlation to the total success/social value score S s 347. The weight for each sub‐score is a
ssigned in the range from 1 to 5. The most important sub‐score (i.e., that sub‐score
or sub‐scores that have the strongest correlation
to the total success/social value score S s 347 and make the largest contribution to the
success/social value score S s 347) is weighted 5, and less important scores
(i.e., that sub‐score or sub‐scores that have the weakest correlation to the total success/soc
ial value score S s 347 and makes the smallest contribution to the success/social value score S s 347) is weighted 1. The weights can be stati
cally applied to each of the sub‐scores, or can be dyna
mically determined and applied based on received fast data. [000118] In this example, weight factor W i is a dynamic factor determined as fast data
is received. Weight factor W i is much lower when the overall success/social
value score S s 347 is lower than 350 and much higher when the overall success/so
cial value score S s 347 is from 351 to 500. [000119] SCF Internal Historical Track Record score S i 348 is also used in the dynamic credit limit determination to quantify and account for past
supplier‐ supply chain finance provider interactions. The interactions can include longevity
of the business relationship, historical payment defaults by the supplier (i.e., payments from the su
pplier to the supply chain finance service provider
in case of recourse), characteristics of the selected
invoices requested for early payment (e.g., amounts, timeframes, etc.), and other supply chain fi
nance internal historical track record items. The SCF Internal Historical Track Record score S i 348 is based on internal supply chain finance
provider’s data that is continually analyzed by the system. I
n some instances, the system determines of sub‐ scores that, in aggregate, make up the SCF Internal
Historical Track Record score S i 348. In FIG. 3, an example is shown where SCF Internal Historical Track
Record score S i 348 is determined based on sub‐sub‐scores 357, 358 (e.g., the sub‐scores s i1 ‐s i5 described below). In FIG. 3, the arrows ext
ending downward from SCF Internal Historical Track Record sc
ore S i 348 represent further sub‐sub‐scores and respective sources 367, 368 of data used to det
ermine the sub‐sub‐scores). In one example of the invention, determining a SCF Internal Historical
Track Record score S i 348 is calculated according to the following relationship: S i = (S i1 *W 1 + S i2 *W 2 + S i3 *W 3 + S i4 *W 4 + S i5 *W 5 )/ Σ(W i ) where S i1 is Longevity of the Account score, calculated
based on longevity history with supply chain finance provider S i2 is Defaults Number/Transactions Number Ratio sc
ore, calculated based on supplier Defaults Number/Transactions Number Ratio historically occurred
with supply chain finance provider funding, S i3 is Total Δ c /Total G e Ratio score, calculated based on supplier Tota
l Δ c /Total G e Ratio historically occurred with supply chain finance provid
er funding, S i4 is Total Δ c /Total (R c + R n ) Ratio score, calculated based on supplier Tot
al Δ c /Total (R c + R n ) Ratio historically occurred with supply chain finance
provider funding, S s5 is Total G c score calculated based on supplier Total G c score historically occurred with supply chain finance provider funding, and W i is a weight factor. [000120] As above, each of the sub‐scores can be weighted
by the supply chain finance provider, and the weighting factors (W 1 , W 2 , etc.) are applied to each of the respective
sub‐scores to determine the overall SCF Internal Historical Track R
ecord score S i 348. As before, weighting the individual sub‐scores that make up the total SCF I
nternal Historical Track Record score S i 348 provides a method of prioritizing sub‐scores and determining
a relative value of that sub‐score’s contribution
to the total SCF Internal Historical Track Record sc
ore S i 348. In one example implementation of the invention, criteria for weighting is selected by anal
yzing statistics and identifying characteristics of each sub‐score’s correlation to the total SCF Int
ernal Historical Track Record score S i 348. The weight for each sub‐score is assigned in the range from
1 to 5. The most important sub‐score (i.e., that
sub‐ score or sub‐scores that have the strongest correla
tion to the total SCF Internal Historical Track Record score S i 348 and make the largest contribution to the S
CF Internal Historical Track Record score S i 348) is weighted 5, and less important scores
(i.e., that sub‐score or sub‐scores that have the
weakest correlation to the total SCF Internal Histori
cal Track Record score S i 348 and makes the smallest contribution to the SCF Internal Historical
Track Record score S i 348) is weighted 1. The weights can be statically applied to each of the su
b‐scores, or can be dynamically determined and applied based on received fast data. [000121] As mentioned briefly above, for the dynamic credit l
imit determination, the Financial Analysis Score 342 and Dilution Prediction component
based on machine learning 333 are optional and can be used when the supply chain finance provi
der has access via integration to the respective data sources. The same is true for the Success/Socia
l Value Score 347. [000122] Other elements (scores, sub‐scores, and sub‐sub‐s
cores) can be reliably used in the systems and methods in accordance with the invention
to produce reliable results. Components (sub‐scores) can be calculated based on fast data
from the different data sources. The better and richer the data sources, the better the results that
are produced by the methods and systems of the invention. Examples of How Methods and Systems of the Invention
Can Be Used Cash planner [000123] One example of how the dynamic credit limit determin
ation can be used is when suppliers need a specific amount of money (e.g., for
payroll, inventory, working capital, etc.), a cash
planner feature enables a user to enter an amount o
f money needed and automatically identifies invoices for early payment that total closest amount
to the amount of money entered by the supplier (up to the amount of the dynamic credit limit). T
his provides a powerful cash management tool. [000124] As shown in FIG. 5, from the invoice page of the
supply chain finance supplier portal, the supplier selects the cash planner button and sel
ects the currency of invoices 505 that the user wishes to review and enters an early payment date 5
10. The supplier enters an amount of money needed in the “How much do you need?” box 515.
The amount needed cannot be greater than the dynamic credit limit 511 (the amount of early paymen
t funding for which the supplier is approved). The user submits the cash planner using the process
button 520, and the invoices 532, 533, 534, 535, 536, 537 are automatically identified for early payme
nt to get to the closest amount of money needed. That is, invoices are identified that total
561 (or are close to) the amount of money needed
515. The supplier can then review the invoices 532
, 533, 534, 535, 536, 537 identified and can de‐ select any of the identified invoices and manually s
elect others, if desired. Once the list of invoice
s is finalized, the supplier submits the list of invoices
571, and the request is processed. In this fashio
n, an automated solution for cash flow planning is prov
ided. Always Pay Me Early [000125] An additional example of how the dynamic credit limi
t determination can be used is when suppliers would like the digital supply chain f
inance method performed automatically (for example, without reviewing a list of invoices for ea
rly payment), the supplier can select an “always pay me early” feature shown in FIG. 6. This fea
ture provides early payment of invoices approved and
scheduled for payment by the buyer up to the amount
of the dynamic credit limit 611. [000126] As shown in FIG. 6, from the invoice page of the
supply chain finance supplier portal, the supplier selects the always pay me early button
603 and selects the currency of issued invoices 605 which the user intends to select for funding.
An “Always Pay Me Early” icon is displayed (e.g
., reference numerals 632, 633) indicating the feature i
s selected. The supplier then enters a frequency of payment 607 (e.g., weekly, bi‐weekly, monthly, q
uarterly). The supplier uses a calendar to choose
a payment date 609 (e.g., monthly on the fifteenth
of each month). The user submits 681 the always pay me early selections, and on the payment date, a
notification and details (via email, for example) of the early payment transactions is provided. The
“always pay me early” feature can be turned on
and off at any time simply by using the button 603
on the interface screen (of FIG. 6). [000127] The systems and methods of the invention allow the
supply chain finance providers to customize and add additional components (scores, s
ub‐scores, sub‐sub‐scores) with identified weighting contributions and additional data sources th
at the supply chain finance provider finds valuable in determining the Dynamic Credit Limit calc
ulations. [000128] By using the systems and methods of the invention,
supply chain finance providers can offer supply chain finance funding to Suppliers
without requiring a Buyer’s guaranty.