Process optimization using integrated digital twins

1 . A computer-implemented method for optimizing processes using integrated digital twins, the method comprising:

receiving, by an integrated digital twin module, an integrated digital twin comprising multiple digital twins, each digital twin comprising a computer-executable model of a real-world system used to execute a portion of a process, within the integrated digital twin, a first digital twin providing output that is used to generate input to a second digital twin;

receiving enterprise data from an enterprise data layer, the enterprise data being provided from a set of real-world systems used to execute the process;

executing, by the integrated digital twin module, one or more simulations of the process using the integrated digital twin and one or more agent-based models based on the enterprise data, at least one agent-based model providing input to the first digital twin;

determining simulation results from the one or more simulations, the simulation results comprising a value of at least one objective function,

wherein the at least one objective function comprises a set of variables, a set of constraints, and a set of weights, and

wherein weights in the set of weights are determined using one or more ML models; and

adjusting one or more parameters of at least one real-world system used to execute the process based on the simulation results, and

wherein adjusting one or more parameters of at least one real-world system comprises changing a parameter value of the at least one real-world system from a first value to a second value, the second value being provided in the simulation results.

2 . The method of claim 1 , further comprising processing at least a portion of real-world data through a signal enhancer to determine a set of patterns, each digital twin being provided based on at least one pattern in the set of patterns, the real-world data representative of historical executions of the process.

3 . The method of claim 1 , wherein each digital twin is provided based on concepts recorded in a knowledge graph representative of relationships between the concepts in a context of the process.

4 . The method of claim 1 , wherein each agent-based model comprises attribute data and behavior data, the attribute data comprising statistical data determined from historical data, the behavior data being determined using a set of machine learning (ML) models.

5 . A system, comprising:

one or more processors; and

a computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for optimizing processes using integrated digital twins, the operations comprising:

receiving, by an integrated digital twin module, an integrated digital twin comprising multiple digital twins, each digital twin comprising a computer-executable model of a real-world system used to execute a portion of a process, within the integrated digital twin, a first digital twin providing output that is used to generate input to a second digital twin;

receiving enterprise data from an enterprise data layer, the enterprise data being provided from a set of real-world systems used to execute the process;

executing, by the integrated digital twin module, one or more simulations of the process using the integrated digital twin and one or more agent-based models based on the enterprise data, at least one agent-based model providing input to the first digital twin;

determining simulation results from the one or more simulations, the simulation results comprising a value of at least one objective function,

wherein the at least one objective function comprises a set of variables, a set of constraints, and a set of weights, and

wherein weights in the set of weights are determined using one or more ML models; and

adjusting one or more parameters of at least one real-world system used to execute the process based on the simulation results, and

wherein adjusting one or more parameters of at least one real-world system comprises changing a parameter value of the at least one real-world system from a first value to a second value, the second value being provided in the simulation results.

6 . The system of claim 5 , wherein operations further comprise processing at least a portion of real-world data through a signal enhancer to determine a set of patterns, each digital twin being provided based on at least one pattern in the set of patterns, the real-world data representative of historical executions of the process.

7 . The system of claim 5 , wherein each digital twin is provided based on concepts recorded in a knowledge graph representative of relationships between the concepts in a context of the process.

8 . The system of claim 5 , wherein each agent-based model comprises attribute data and behavior data, the attribute data comprising statistical data determined from historical data, the behavior data being determined using a set of machine learning (ML) models.

9 . A non-transitory computer-readable storage media coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for optimizing processes using integrated digital twins, the operations comprising:

receiving, by an integrated digital twin module, an integrated digital twin comprising multiple digital twins, each digital twin comprising a computer-executable model of a real-world system used to execute a portion of a process, within the integrated digital twin, a first digital twin providing output that is used to generate input to a second digital twin;

receiving enterprise data from an enterprise data layer, the enterprise data being provided from a set of real-world systems used to execute the process;

executing, by the integrated digital twin module, one or more simulations of the process using the integrated digital twin and one or more agent-based models based on the enterprise data, at least one agent-based model providing input to the first digital twin;

determining simulation results from the one or more simulations, the simulation results comprising a value of at least one objective function,

wherein the at least one objective function comprises a set of variables, a set of constraints, and a set of weights, and

wherein weights in the set of weights are determined using one or more ML models; and

adjusting one or more parameters of at least one real-world system used to execute the process based on the simulation results, and

wherein adjusting one or more parameters of at least one real-world system comprises changing a parameter value of the at least one real-world system from a first value to a second value, the second value being provided in the simulation results.

10 . The computer-readable storage media of claim 9 , wherein operations further comprise processing at least a portion of real-world data through a signal enhancer to determine a set of patterns, each digital twin being provided based on at least one pattern in the set of patterns, the real-world data representative of historical executions of the process.

11 . The computer-readable storage media of claim 9 , wherein each digital twin is provided based on concepts recorded in a knowledge graph representative of relationships between the concepts in a context of the process.

12 . The computer-readable storage media of claim 9 , wherein each agent-based model comprises attribute data and behavior data, the attribute data comprising statistical data determined from historical data, the behavior data being determined using a set of machine learning (ML) models.

13 . The method of claim 1 , wherein executing the one or more simulations by the integrated digital twin module comprises interaction with one or more devices through an interface module, wherein for the one or more simulations:

receiving in a user communication, the objective function and constraints that are to be applied using the integrated digital twin; and

outputting the integrated digital twin a result of the one or more simulations to the user.

14 . The system of claim 5 , wherein executing the one or more simulations by the integrated digital twin module comprises interaction with one or more devices through an interface module, wherein for the one or more simulations, a user communicates the objective function and constraints that are to be applied using the integrated digital twin, and wherein the integrated digital twin communicates a result of the one or more simulations to the user.