IP Library Granted Patent US 12683868
Granted Patent B1
US 12683868 · App. 18/911,039 · Granted Jul 14, 2026

Systems and methods for risk simulation based transaction routing

Inventors: Thomas Booker (New York, NY); Misha Van Beek (New York, NY); Bingxu Chen (New York, NY); David Greenberg (New York, NY)
Assignee: BlackRock Finance, Inc.
H04L41/145H04L41/0836
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12683868
App. No.
18/911,039
Granted
Jul 14, 2026
Kind
B1
Abstract

Embodiments described herein further provide a routing server for routing incoming transaction requests based on a stochastic simulation of future risk level, e.g., demand flows such as required network resource, and/or the like.

Claims (69)

1 . A routing system for routing a network access request based on a risk level, the routing system comprising:

a communication interface receiving the network access request comprising information relating to a network application including a plurality of application attributes;

a memory that stores a plurality of processor-executable instructions;

one or more processors that reads and executes the plurality of processor-executable instructions from the memory to perform operations comprising:

decomposing the network application into a set of discrete execution processes, each of which is subject to a respective stochastic entry time and a respective stochastic exit time in and out of the network application, and a respective stochastic waiting time having a hazard-based duration, respectively,

wherein the respective stochastic entry time or the respective stochastic exit time takes a form of a hazard rate model subject to an exponential process with time-varying coefficients;

a hardware-based stochastic simulator that:

conducting a stochastic simulation of entry dynamics or exit dynamics of the set of discrete execution processes based at least in part on simulating a stochastic entry intensity rate or an exit intensity rate according to a hazard rate model, based on the plurality of application attributes; and

generating, from simulation outputs, a predicted risk level associated with a forecasted demand flow from the network application over a future period;

wherein the one or more processors further perform operations comprising generating a routing decision for the network access request based on the predicted risk level.

2 . The routing system of claim 1 , wherein the operations further comprise:

causing a display of a visualization of the predicted risk level over a period of time on a user interface;

receiving, via the user interface, a user input that continuously move a marker on the user interface to indicate a changed time period; and

causing an updated display of a visualization of an updated predicted risk level over the changed time period on the user interface.

3 . The routing system of claim 1 , wherein the operations of generating the routing decision further comprising:

declining the network access request when the predicted risk level indicates a high risk; and

blocking future network access requests from the requestor server or relating to the target server.

4 . The routing system of claim 1 , wherein the operations of generating the routing decision further comprising:

routing, by a routing lookup table, the network access request to a destination server when the predicted risk level indicates an acceptable risk.

5 . The routing system of claim 1 , wherein the network access request comprises a transaction request from a requester server to initiate a transaction with a destination server, and wherein the network application comprises an investment vehicle including a plurality of vehicle attributes.

6 . The routing system of claim 5 , wherein the plurality of vehicle attributes include any combination of a strategy of the investment vehicle, a total vehicle life, an investment period length, a total commitment.

7 . The routing system of claim 5 , wherein the respective stochastic entry time or the respective stochastic exit time takes a form of a hazard rate model subject to an exponential process with time-varying coefficients.

8 . The routing system of claim 5 , further comprising:

calibrating, using a regression model, one or more entry intensity parameters or exit intensity parameters including a lagged intensity value, an age of the investment vehicle, and a risk-free rate and a market return based on historical capital call intensity data or historical distribution intensity data.

9 . The routing system of claim 5 , wherein the stochastic simulation includes simulating a capital call intensity rate or a distribution intensity rate for the investment vehicle, based at least in part on:

(i) obtaining a set of economic factors that drive capital calls, distributions or returns of a set of discrete portfolio investment opportunities decomposed from the investment vehicle;

(ii) generating the simulated capital call intensity rate or the simulated distribution intensity rate based on the set of economic factors;

(iii) generating simulated returns from the set of discrete portfolio investment opportunities by drawing the set of discrete portfolio investment opportunities from a Gaussian distribution;

(iv) generating a simulated entry time and a simulated exit time for each discrete portfolio investment opportunity based on transition probabilities computed from the simulated capital call intensity rate and the simulated distribution intensity rate;

(v) aggregating simulated activities of the set of discrete portfolio investment opportunities to obtain a capital drawdown, a distribution and a return value for the private equity investment vehicle; and

(vi) generating a distribution of state variable trajectory corresponding to investment states of the set of discrete portfolio investment opportunities.

10 . The routing system of claim 9 , wherein the simulated entry time and the simulated exit time are generated by sequentially simulating a set of investment states of each discrete portfolio investment opportunity,

wherein the set of investment states includes an uninvested state that transits to an invested state according to a first transition probability computed based on the simulated capital call intensity rate, and

wherein the invested state transits to a divested state according to a second transition probability computed based on the simulated distribution intensity rate.

11 . A method for routing a network access request based on a risk level, the method comprising:

receiving by a communication interface, the network access request comprising information relating to a network application including a plurality of application attributes;

decomposing, by a processor, the network application into a set of discrete execution processes, each of which is subject to a respective stochastic entry time and a respective stochastic exit time in and out of the network application, and a respective stochastic waiting time having a hazard-based duration, respectively,

wherein the respective stochastic entry time or the respective stochastic exit time takes a form of a hazard rate model subject to an exponential process with time-varying coefficients;

conducting, by a hardware-based stochastic simulator, a stochastic simulation of entry dynamics or exit dynamics of the set of discrete execution processes based at least in part on simulating a stochastic entry intensity rate or an exit intensity rate according to a hazard rate model, based on the plurality of application attributes;

generating, from simulation outputs, a predicted risk level associated with a forecasted demand flow from the network application over a future period; and

generating a routing decision for the network access request based on the predicted risk level.

12 . The method of claim 11 , further comprising:

causing a display of a visualization of the predicted risk level over a period of time on a user interface;

receiving, via the user interface, a user input that continuously move a marker on the user interface to indicate a changed time period; and

causing an updated display of a visualization of an updated predicted risk level over the changed time period on the user interface.

13 . The method of claim 11 , wherein the generating the routing decision further comprises:

declining the network access request when the predicted risk level indicates a high risk; and

blocking future network access requests from the requestor server or relating to the target server.

14 . The method of claim 11 , wherein the generating the routing decision further comprises:

routing, by a routing lookup table, the network access request to a destination server when the predicted risk level indicates an acceptable risk.

15 . The method of claim 1 , wherein the network access request comprises a transaction request from a requester server to initiate a transaction with a destination server, and wherein the network application comprises an investment vehicle including a plurality of vehicle attributes.

16 . The method of claim 15 , wherein the plurality of vehicle attributes include any combination of a strategy of the investment vehicle, a total vehicle life, an investment period length, a total commitment.

17 . The method of claim 15 , wherein the respective stochastic entry time or the respective stochastic exit time takes a form of a hazard rate model subject to an exponential process with time-varying coefficients.

18 . The method of claim 15 , further comprising:

calibrating, using a regression model, one or more entry intensity parameters or exit intensity parameters including a lagged intensity value, an age of the investment vehicle, and a risk-free rate and a market return based on historical capital call intensity data or historical distribution intensity data.

19 . The method of claim 15 , wherein the stochastic simulation includes simulating a capital call intensity rate or a distribution intensity rate for the investment vehicle, based at least in part on:

(i) obtaining a set of economic factors that drive capital calls, distributions or returns of a set of discrete portfolio investment opportunities decomposed from the investment vehicle;

(ii) generating the simulated capital call intensity rate or the simulated distribution intensity rate based on the set of economic factors;

(iii) generating simulated returns from the set of discrete portfolio investment opportunities by drawing the set of discrete portfolio investment opportunities from a Gaussian distribution;

(iv) generating a simulated entry time and a simulated exit time for each discrete portfolio investment opportunity based on transition probabilities computed from the simulated capital call intensity rate and the simulated distribution intensity rate;

(v) aggregating simulated activities of the set of discrete portfolio investment opportunities to obtain a capital drawdown, a distribution and a return value for the private equity investment vehicle; and

(vi) generating a distribution of state variable trajectory corresponding to investment states of the set of discrete portfolio investment opportunities.

20 . A non-transitory processor-readable storage medium storing a plurality of processor-executable instructions for analyzing or constructing an asset portfolio by simulating private equity cash flows, the processor-executable instructions being executed by a processor to perform operations comprising:

receiving, a communication interface, the network access request comprising information relating to a network application including a plurality of application attributes;

decomposing, by a processor, the network application into a set of discrete execution processes, each of which is subject to a respective stochastic entry time and a respective stochastic exit time in and out of the network application, and a respective stochastic waiting time having a hazard-based duration, respectively,

wherein the respective stochastic entry time or the respective stochastic exit time takes a form of a hazard rate model subject to an exponential process with time-varying coefficients;

conducting, by a hardware-based stochastic simulator, a stochastic simulation of entry dynamics or exit dynamics of the set of discrete execution processes based at least in part on simulating a stochastic entry intensity rate or an exit intensity rate according to a hazard rate model, based on the plurality of application attributes;

generating, from simulation outputs, a predicted risk level associated with a forecasted demand flow from the network application over a future period; and

generating a routing decision for the network access request based on the predicted risk level.