System maintenance using unified cognitive root cause analysis for multiple domains
A computer-implemented method, system, and non-transitory computer program product for maintaining a system. A domain of the system is identified. Problem information identifying a problem in the system is captured. Key performance indicators are obtained from a historical model database for the identified domain. An unsupervised model is applied to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system. A linear complexity model is used to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem. A real-time objective function is used to select a solution to the problem from the potential solutions. A maintenance operation to be performed on the system to implement the solution and fix the problem is identified.
1. A computer-implemented method of maintaining a system, comprising:
identifying a domain of the system;
capturing problem information identifying a problem in the system;
obtaining key performance indicators from a historical model database for the identified domain;
applying an unsupervised model to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system;
determining time-cost complexity and resource-cost complexity of the historical solutions;
using a linear complexity model to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem;
using a real-time objective function to select a solution to the problem from the potential solutions; and
identifying a maintenance operation to be performed on the system to implement the solution and fix the problem.
2. The computer-implemented method of claim 1 , wherein the domain is selected from the group of domains consisting of information technology, automotive, healthcare, and aerospace.
3. The computer-implemented method of claim 1 , wherein capturing the problem information comprises capturing real-time performance metrics of the system.
4. The computer-implemented method of claim 1 further comprising generating the real-time objective function using a neural network.
5. The computer-implemented method of claim 1 further comprising using a non-linear constraint model to select the solution to the problem, wherein the non-linear constraint model applies at least one constraint to selecting the solution to the problem that is selected from the group of constraints consisting of: part availability for the solution, skill level of available personnel for implementing the solution, process limitations, and design limitations of the system.
6. The computer-implemented method of claim 1 , wherein the key performance indicators for the historical model database are generated by an offline process using an ensemble of root cause analysis methods.
7. The computer-implemented method of claim 1 further comprising performing the maintenance operation on the system to fix the problem.
8. An apparatus for maintaining a system, comprising:
an operator interface configured to receive a selected domain of the system and operator provided problem information identifying a problem in the system;
a historical model database comprising key performance indicators for the selected domain;
an unsupervised model configured to be applied to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system;
a linear complexity model configured to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem; and
an objective function generator configured to generate a real-time objective function to select a solution to the problem from the potential solutions, wherein the solution identifies a maintenance operation to be performed on the system to implement the solution and fix the problem.
9. The apparatus of claim 8 , wherein the wherein the domain is selected from the group of domains consisting of information technology, automotive, healthcare, and aerospace.
10. The apparatus of claim 8 , wherein the problem information further comprises real-time performance metrics for the system.
11. The apparatus of claim 8 , wherein the objective function generator comprises a neural network.
12. The apparatus of claim 8 further comprising a non-linear constraint model configured to apply at least one constraint to selecting the solution to the problem that is selected from the group of constraints consisting of: part availability for the solution, skill level of available personnel for implementing the solution, process limitations, and design limitations of the system.
13. The apparatus of claim 8 , wherein the key performance indicators for the historical model database are generated by an offline process using an ensemble of root cause analysis methods.
14. A non-transitory computer-readable storage medium storing program code, which when executed by a processor, performs a computer-implemented method of maintaining a system, comprising:
identifying a domain of the system;
capturing problem information identifying a problem in the system;
obtaining key performance indicators from a historical model database for the identified domain;
applying an unsupervised model to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system;
determining time-cost complexity and resource-cost complexity of the historical solutions;
using a linear complexity model to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem;
using a real-time objective function to select a solution to the problem from the potential solutions; and
identifying a maintenance operation to be performed on the system to implement the solution and fix the problem.
15. The non-transitory computer-readable storage medium of claim 14 , wherein the domain is selected from the group of domains consisting of information technology, automotive, healthcare, and aerospace.
16. The non-transitory computer-readable storage medium of claim 14 , wherein the problem information comprises real-time performance metrics.
17. The non-transitory computer-readable storage medium of claim 14 , wherein the key performance indicators for the historical model database are generated by an offline process using an ensemble of root cause analysis methods.
18. The non-transitory computer-readable storage medium of claim 14 , wherein the real-time objective function is generated using a neural network.
19. The non-transitory computer-readable storage medium of claim 14 , wherein the program code, when executed by a processor, further performs the step of using a non-linear constraint model to select the solution to the problem, wherein the non-linear constraint model applies at least one constraint to selecting the solution to the problem that is selected from the group of constraints consisting of: part availability for the solution, skill level of available personnel for implementing the solution, process limitations, and design limitations of the system.
20. The non-transitory computer-readable storage medium of claim 14 , wherein the program code, when executed by a processor, further performs the step of performing the maintenance operation on the system to fix the problem.