Systems and improvement methods for interactive relationship graph to adjust performance metrics

1 . A system comprising:

a processor, configured to:

read parameter information corresponding to information regarding a plurality of evaluation indexes in an operational process;

calculate a first correlation in which a correlation between the evaluation indexes is expressed by a correlation coefficient;

create a causal relationship graph, in which the evaluation indexes are set as nodes and the correlation between the evaluation indexes is represented by a directed edge having a first evaluation index of a higher rank in an order information as a start point and a second evaluation index of a lower rank in order information as an end point, based on the first correlation and the order information corresponding to information regarding priorities of the evaluation indexes;

present, via a display device, the causal relationship graph to a user;

receive, via an input device, a change to a direction of directed edge in the causal relationship graph by the user;

update the order information based on the change to the direction of directed edge in the causal relationship graph by the user;

subsequent to updating the order information, perform a simple regression analysis using a third evaluation index of a lower rank in the updated order information as an explanatory variable, and a fourth evaluation index of a higher rank in the updated order information as an objective variable;

calculate a value based on a slope and a sign of a linear term of a regression expression of the simple regression analysis as an improvement effect;

update the causal relationship graph based on the updated order information and the improvement effect; and

display, via the display device, the updated causal relationship graph based on the updated order information and the improvement effect.

2 . The system according to claim 1 , wherein the processor is configured to:

display, via the display device on the causal relationship graph, at least one of the evaluation indexes of which the correlation coefficient is greater than or equal to a predetermined threshold.

3 . The system according to claim 2 , wherein the processor is configured to:

calculate the correlation coefficient by linear approximation.

4 . The system according to claim 1 , wherein the processor is configured to:

extract, as a difference evaluation index, at least one of the evaluation indexes index-related to the change to the causal relationship graph;

create a plurality of pieces of provisional order information in which an order of the difference evaluation index is changed in the order information; and

set, as the updated order information, the provisional order information in which a temporary causal relationship graph, that is created based on the provisional order information and the parameter information, coincide with the causal relationship graph after the change.

5 . The system according to claim 1 , wherein the the processor is configured to:

receive, via the input device, an operation of the user changing the direction of directed edge in the causal relationship graph.

6 . The system according to claim 1 , wherein based on selection of a node of the nodes by the user, the processor is configured to:

present the improvement effect related to the selected node to the user.

7 . The system according to claim 1 , wherein the the processor is configured to:

multiply a quantity in the improvement effect by a predetermined magnification a based on an operation of the user via the input device; and

display, via the display device, a result of multiplying the quantity by the predetermined magnification.

8 . The system according to claim 1 , wherein the evaluation index includes at least one of man-hours or cost.

9 . The system according to claim 1 , wherein the processor is configured to:

hide one or more of the evaluation indexes based on an operation of the user via the input device.

10 . A method, comprising:

reading, by a processor, parameter information corresponding to information regarding a plurality of evaluation indexes in an operational process;

calculating, by the processor, a first correlation in which a correlation between the evaluation indexes is expressed by a correlation coefficient;

creating, by the processor, a causal relationship graph, in which the evaluation indexes are set as nodes and the correlation between the evaluation indexes is represented by a directed edge having a first evaluation index of a higher rank in an order information as a start point and a second evaluation index of a lower rank in order information as an end point, based on the first correlation and the order information corresponding to information regarding priorities of the evaluation indexes;

presenting, by the processor via a display device, the causal relationship graph to a user;

receiving, by the processor via an input device, a change to a direction of directed edge in the causal relationship graph by the user;

updating, by the processor, the order information based on the change to the direction of directed edge in the causal relationship graph by the user;

subsequent to updating the order information, performing, by the processor, a simple regression analysis using, a third evaluation index of a lower rank in the updated order information as an explanatory variable, and a fourth evaluation index of a higher rank in the updated order information as an objective variable;

calculating, by the processor, a value based on a slope and a sign of a linear term of a regression expression of the simple regression analysis as an improvement effect;

updating, by the processor, the causal relationship graph based on the updated order information and the improvement effect; and

displaying, by the processor via the display device, the updated causal relationship graph based on the updated order information and the improvement effect.