Systems and method for automatic transaction routing and execution
According to one aspect, systems and methods are provided that intelligently automate payment routing. According to various embodiments, the intelligent routing can be augmented with machine learning models that generate an optimal routing of a transaction, wherein for example, the optimal routing can be based on a unique set of circumstances for each transaction. In further embodiments, machine learning models accept payment processing information and circumstance (e.g., payment modality, store location, online payment location, source IP address, payor IP address, merchant IP address, physical presence, intermediary information, merchant information, payment brand, supplier data, among a host of other options) and based on the trained model can generate an optimal path for any particular transaction.
1 . An intelligent routing system, comprising:
at least one processor operatively connected to a memory, the at least one processor when executing configured to:
present an interface to a customer comprising one or more fields and receiving information associated with an initiation of a transaction including identification data, payment information, and contact information for the customer;
accept as input to at least one machine learning model the information associated with an initiation of the transaction;
evaluate with the at least one machine learning model a plurality of processing routes for the transaction, the evaluation comprising:
for a first processing route of the plurality of processing routes, determine a first computational burden associated with executing the transaction, the first processing route comprising a first set of system routing options including (1) one or more intermediary systems, (2) requirements associated with the intermediary systems, and (3) any triggers associated with additional computational burdens including analysis of any trigger that increases security information requested, the first computational burden being determined based at least in part on the requirements associated with the intermediary systems and the triggers caused by executing the transaction according to the first processing route, including any trigger that increases security information requested;
for a second processing route of the plurality of processing routes, determine a second computational burden associated with executing the same transaction, the second processing route comprising a second set of system routing options including (1) one or more intermediary systems, (2) requirements associated with the intermediary systems, (3) and any triggers associated with additional computational burdens, the second computational burden being determined based at least in part on the requirements associated with the intermediary systems and the triggers caused by executing the transaction according to the second processing route;
evaluate probabilities of executing the transaction successfully associated with the first processing route and second processing route;
compare the first computational burden with the second computational burden;
based on a first probability of executing the transaction successfully with the first processing route meeting a threshold and the first computational burden associated with the first processing route being less than the second computational burden associated with the second processing route, execute the transaction using the first processing route; and
based on a second probability of executing the transaction successfully with the second processing route meeting the threshold and the second computational burden associated with the second processing route being less than the first computational burden associated with the first processing route, execute the transaction using the second processing route.
2 . The system of claim 1 , wherein the at least one processor is further configured to analyze security options for the plurality of processing routes to determine the at least one route having a highest probability of success.
3 . The system of claim 2 , wherein the at least one processor is further configured to determine the at least one route having the highest probability of success based on evaluation of enhanced security protocols for each processor, payment method, intermediary, and acquirer system of the plurality of routes.
4 . The system of claim 1 , wherein the at least one processor is further configured to determine enhanced security protocols negatively impact successful operation.
5 . The system of claim 4 , wherein the at least one processor is further configured to select an alternative routing to eliminate enhanced security protocols.
6 . The system of claim 1 , wherein the at least one processor is configured to:
determine short term analysis of the plurality of processing routes disagrees with the at least one machine learning model;
dynamically weight the short term analysis such that selection and execution of the transaction using the at least one processing route is based on the short term analysis.
7 . The system of claim 6 , wherein the at least one processor is configured to adjust the weighting of the short term analysis based on the observed execution of the at least one processing route.
8 . The system of claim 1 , wherein the at least one processor is configured to:
based on a plurality of output probabilities meeting the threshold, select and execute the transaction using a respective one of a plurality of processing routes having a greatest probability of returning a successful transaction.
9 . The system of claim 1 , wherein the at least one machine learning model is trained on historical payment information.
10 . The system of claim 9 , wherein the historical payment information includes attributes specifying at least one or more of geographic information, payment user's device, connection details, and an aggregated reputation of the transaction.
11 . The system of claim 1 , wherein the at least one machine learning model is trained to output the probability of executing the transaction successfully based on an evaluation of a plurality of processing systems included in a processing route.
12 . The system of claim 1 , wherein the at least one machine learning model is trained to output the probability of executing the transaction successfully based on prior knowledge of a payment user's actions.
13 . The system of claim 1 , wherein the at least one machine learning model is trained to output the probability a payment user will respond positively to an enhanced security challenge.
14 . The system of claim 1 , wherein the at least one machine learning model is trained to evaluate timing of presentation of the information associated with the transaction to output the probability of executing the transaction in a specific time frame.
15 . The system of claim 1 , wherein the at least one processor is configured to generate a plurality of candidate routes for executing the transaction.
16 . The system of claim 9 , wherein the at least one processor is configured to evaluate a respective route based on the probability that a respective merchant processing triggers enhanced security requirements.
17 . A computer implemented method for intelligent routing, the method comprising:
presenting an interface to a customer comprising one or more fields and receiving information associated with an initiation of a transaction including identification data, payment information, and contact information for the customer;
accepting, by at least one processor, information associated with the initiation of the transaction as input to at least one machine learning model the information associated with an initiation of the transaction;
evaluating, with the at least one machine learning model executed by the at least one processor, a plurality of processing routes for the transaction, the evaluation comprising:
for a first processing route of the plurality of processing routes, determining a first computational burden associated with executing the transaction, the first processing route comprising a first set of system routing options including (1) one or more intermediary systems, and a first outcome system, (2) requirements associated with first participant systems, and (3) any triggers associated with additional computational burdens including analysis of any trigger that increases if security information requested, the first computational burden being determined based at least in part on the requirements associated with the intermediary systems and the triggers caused by executing the transaction according to the first processing route, including any trigger that increases security information requested;
for a second processing route of the plurality of processing routes, determining a second computational burden associated with executing the same transaction, the second processing route comprising a second set of system routing options including (1) one or more intermediary systems, (2) requirements associated with the intermediary systems, and (3) any triggers associated with additional computational burdens, the second computational burden being determined based at least in part on the requirements associated with the intermediary systems and the triggers caused by executing the transaction according to the second processing route; and
evaluating by the at least one machine learning model probabilities of executing the transaction successfully associated with the first processing route and second processing route;
compare the first computational burden and the second computational burden;
executing, by the at least one processing, the transaction using the first processing route based on a first probability of executing the transaction successfully with the first processing route meeting a threshold and the first computational burden associated with the first processing route being less than the second computational burden associated with the second processing route; and
executing, by the at least one processor, the transaction using the second processing route based on a second probability of executing the transaction successfully with the second processing route meeting a threshold and the second computational burden associated with the second processing route being less than the first computational burden associated with the first processing route.
18 . The method of claim 17 , wherein the method further comprises analyzing security options for the plurality of processing routes to determine the at least one route having a highest probability of success.
19 . The method of claim 18 , wherein the method further comprises determining the at least one route having the highest probability of success based on evaluation of enhanced security protocols for each processor, payment method, intermediary, and acquirer system of the plurality of routes.
20 . The method of claim 17 , wherein the method further comprises determining enhanced security protocols negatively impact successful operation.
21 . The method of claim 20 , wherein the method further comprises selecting an alternative routing to eliminate enhanced security protocols.
22 . The method of claim 17 , wherein the method further comprises:
determining short term analysis of the plurality of processing routes disagrees with the at least one machine learning model; and
dynamically weighting the short term analysis such that selecting and triggering of the transaction using the at least one processing route is based on the short term analysis.
23 . The method of claim 22 , wherein the method further comprises adjusting the weighting of the short term analysis based on the observed execution of the at least one processing route.