Systems and methods for multi-objective optimizations with live updates

• US 10,474,952 B2
• Filed: 09/08/2015
• Issued: 11/12/2019
• Est. Priority Date: 09/08/2015
• Status: Active Grant

• Current Assignee: Aerospace Corporation
• Original Assignee: Aerospace Corporation
• Inventors: Thompson, Timothy Guy
• Primary Examiner(s): Nilsson, Eric
• Application Number:
  US14/847,424
• Publication Number:
  US 20170068890A1
• Time in Patent Office: 1,526 Days
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

Systems and methods are provided for generating an initial optimized baseline solution to a multi-objective problem. As the baseline solution is implemented, live (e.g., real-time or near real-time) data associated with one or more parameters may be received and compared to expectations of those parameters with the implementation of the initial optimized solution. If a deviation is detected between the expectation of the time progression of the parameters and live data associated with the parameter, then that deviation may be compared to a threshold. If the deviation meets a threshold condition, then an irregular operation may be declared and a new baseline solution may be implemented. The new baseline solution may be obtained as a re-optimized solution.

Claims

1. A method, comprising:

receiving, by one or more processors from a multi-objective optimization system, a first multi-objective optimized solution, wherein the first multi-objective optimized solution is based at least in part on a first set of decision variables;

providing, by the one or more processors, the first multi-objective solution as a baseline for implementation;

identifying, by the one or more processors, a parameter associated with the first multi-objective optimization;

receiving, by the one or more processors, a prediction of the parameter;

receiving, by the one or more processors, live data associated with the parameter;

determining, by the one or more processors, that an irregular operation has occurred by comparing the prediction of the parameter with the live data associated with the parameter;

receiving, by the one or more processors and responsive to determining that the irregular operation has occurred, a second multi-objective optimized solution; and

providing, by the one or more processors, the second multi-objective optimized solution as the baseline, wherein the second multi-objective optimized solution is based at least in part on a second set of decision variables, wherein the second set of decision variables comprise a different number of decision variables than the first set of decision variables.

2. The method of claim 1, further comprising:

determining, by the one or more processors, the second set of decision variables, wherein the second set of decision variables have fewer decision variables than the first set of decision variables; and

providing, by the one or more processors, to the multi-objective optimization system, the second set of decision variables.

3. The method of claim 1, wherein the second set of decision variables do not include a particular decision variable from the first set of decision variables, wherein an event associated with the particular decision variable has already transpired.

4. The method of claim 1, wherein the parameter comprises at least one of the first set of decision variables.

5. The method of claim 1, wherein the parameter is derived based at least in part on one or more of the first set of decision variables.

6. The method of claim 1, wherein receiving the first multi-objective optimized solution comprises:

evaluating one or more first potential solutions using one or more constraint models, wherein each of the one or more potential solutions include the first set of decision variables;

evaluating the one or more first potential solutions using one or more objective functions;

performing one or more evolutionary operations on the one or more first potential solutions to generate one or more second potential solutions, wherein the one or more second potential solutions include the first set of decision variables;

identifying a plurality of pareto-optimized solutions from the one or more second potential solutions; and

selecting the multi-objective optimized solution from the plurality of pareto-optimized solutions.

7. The method of claim 1, wherein receiving the live data associated with the parameter comprises requesting the live data from one or more live data sources and receiving the live data from the one or more live data sources.

8. The method of claim 1, wherein determining, by the one or more processors, that an irregular operation has occurred comprises:

identifying a difference between the live data associated with the parameter and the prediction of the parameter; and

determining that the difference is greater than a threshold value.

9. The method of claim 1, wherein receiving the second multi-objective optimized solution comprises:

providing, to the multi-objective optimization system, a second set of decision variables;

providing, to the multi-objective optimization system, one or more objective models; and

initiating a re-optimization at the multi-objective optimization system.

10. The method of claim 1, wherein providing the second multi-objective optimized solution as the baseline comprises:

receiving a plurality of pareto-optimized solutions, wherein the plurality of pareto-optimized solutions include the second multi-objective optimized solution;

presenting two or more of the plurality of pareto-optimized solutions including the second multi-objective optimized solution;

receiving an indication of selection of the second multi-objective optimized solution from the plurality of pareto-optimized solutions; and

indicating the second multi-objective optimized solution as selected from the plurality of pareto-optimized solutions.

11. A system, comprising:

a memory that stores computer-executable instructions;

at least one processor configured to access the memory, wherein the at least one processor is further configured to execute the computer-executable instructions to;

provide the first multi-objective solution as a baseline for implementation;

identify a first parameter associated with the first multi-objective optimization;

receive a prediction of the first parameter;

receive live data associated with the first parameter;

determine that an irregular operation has occurred by comparing the prediction of the first parameter with the live data associated with the first parameter;

receive, responsive to determining that the irregular operation has occurred, a second multi-objective optimized solution; and

provide the second multi-objective optimized solution as the baseline, wherein the second multi-objective optimized solution is based at least in part on a second set of decision variables, wherein the second set of decision variables comprise a different number of decision variables than the first set of decision variables.

12. The system of claim 11, further wherein the at least one processor is further configured to execute the computer-executable instructions to:

determine the second set of decision variables, wherein the second set of decision variables have fewer decision variables than the first set of decision variables; and

provide to a multi-objective optimization system, the second set of decision variables.

13. The system of claim 11, wherein the second set of decision variables do not include a particular decision variables from the first set of decision variables, wherein an event associated with the particular decision variable has already transpired.

14. The system of claim 11, wherein the first parameter comprises at least one of the first set of decision variables.

15. The system of claim 11, wherein the first parameter is based at least in part on one or more of the first set of decision variables.

16. The system of claim 11, wherein the at least one processor is further configured to execute the computer-executable instructions to:

evaluate one or more first potential solutions using one or more constraint models, wherein each of the one or more potential solutions include the first set of decision variables;

evaluate the one or more first potential solutions using one or more objective functions;

perform one or more evolutionary operations on the one or more first potential solutions to generate one or more second potential solutions, wherein the one or more second potential solutions include the first set of decision variables;

identify a plurality of pareto-optimized solutions from the one or more second potential solutions; and

select the multi-objective optimized solution from the plurality of pareto-optimized solutions.

17. The system of claim 11, wherein the at least one processor is configured to receive the live data associated with the first parameter comprises the at least one processor is further configured to execute the computer-executable instructions to request the live data from one or more live data sources and receive the live data from the one or more live data sources.

18. The system of claim 11, wherein the at least one processor is configured to determine that an irregular operation has occurred comprises the at least one processor is further configured to execute the computer-executable instructions to:

identify a difference between the live data associated with the first parameter and the prediction of the first parameter; and

determine that the difference is greater than a threshold value.

19. The system of claim 11, wherein the at least one processor is configured to receive the second multi-objective optimized solution comprises the at least one processor is further configured to execute the computer-executable instructions to:

provide, to a multi-objective optimization system, a second set of decision variables;

provide, to the multi-objective optimization system, one or more objective models; and

initiate a re-optimization at the multi-objective optimization system.

20. The system of claim 11, wherein to provide the second multi-objective optimized solution as the baseline comprises the at least one processor is further configured to execute the computer-executable instructions to:

receive a plurality of pareto-optimized solutions, wherein the plurality of pareto-optimized solutions include the second multi-objective optimized solution;

present two or more of the plurality of pareto-optimized solutions including the second multi-objective optimized solution;

receive an indication of selection of the second multi-objective optimized solution from the plurality of pareto-optimized solutions; and

indicate the second multi-objective optimized solution as selected from the plurality of pareto-optimized solutions.