Improvement system, and improvement method

Abstract

To reflect editing of a causal relation graph on order information.SOLUTION: An improvement system 10 comprises: a first correlation calculation unit that reads parameter information on a plurality of evaluation indices in a business process, and calculates first correlation representing the correlation between the evaluation indices with a correlation coefficient; a graph generation unit that generates a causal relation graph indicating the correlation between the evaluation indices with a link on the basis of the first correlation and information on the order of the plurality of evaluation indices; an interface unit that presents the causal relation graph to a user, and receives a change to the causal relation graph made by the user; and an order calculation unit that extracts an evaluation index related to the change to the causal relation graph made by the user as a differential evaluation index, generates a plurality of kinds of provisional order information obtained by changing the order of the differential evaluation index in the order information, and on the basis of the provisional order information and the parameter information, takes the provisional order information in which a temporary causal relation graph generated by the graph generation unit matches the causal relation graph after the change made by the user, as order information after update.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an improvement system and method. [Background technology]

[0002] In software development projects, improvement measures are sometimes planned based on data obtained from the project. To plan measures efficiently, a causal graph is used, in which project events are treated as nodes and directed lines connect the cause nodes to the result nodes. A causal graph uses nodes and directed lines to represent relationships such as "when one event occurs, another event is more likely to occur." For example, by tracing the directed lines connected to important events such as "increasing project profits" in reverse, planners can find the measures they need to take. Patent Document 1 discloses a causal relationship display system that includes a graph generation unit that generates a directed graph based on causal relationship information, which is information indicating the causal relationships between multiple elements, each of which is a cause or a result, and that generates a directed graph consisting of multiple nodes corresponding to the multiple elements and multiple edges corresponding to the causal relationships between the multiple elements, and a UI (User Interface) control unit that displays output information including the generated directed graph, in which in the directed graph, a first direction, which is either horizontal or vertical, is an x-direction having a +x direction and a -x direction, a second direction perpendicular to the first direction is a y-direction having a +y direction and a -y direction, and some line segments of two or more edges connecting one or more nodes to one or more other nodes are allowed to overlap. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-098387 Summary of the Invention [Problem to be solved by the invention]

[0004] In the invention described in Patent Document 1, edits to the causal relationship graph cannot be reflected in the order information. [Means for solving the problem]

[0005] An improvement system according to a first aspect of the present invention comprises: a first correlation calculation unit that reads parameter information, which is information about a plurality of evaluation indexes in a business process, and calculates a first correlation, which expresses the correlations between the evaluation indexes as correlation coefficients; a graph generation unit that creates a causal relationship graph, with the evaluation indexes as nodes and the correlations between the evaluation indexes as links, based on the first correlation and order information, which is information about the priorities of the plurality of evaluation indexes; an interface unit that presents the causal relationship graph to a user and accepts changes to the causal relationship graph by the user; and an order calculation unit that extracts the evaluation indexes related to the changes to the causal relationship graph by the user as difference evaluation indexes, creates multiple sets of tentative order information by changing the order of the difference evaluation indexes in the order information, and sets as the updated order information the tentative order information, which is the temporary causal relationship graph created by the graph generation unit based on the tentative order information and the parameter information, that matches the causal relationship graph after the change by the user. an order calculation step of updating the temporary order information created by the graph generation step based on the temporary order information and the parameter information, and determining as the updated order information the temporary order information created by the graph generation step that the temporary order information created by the graph generation step matches the causal order graph changed by the user; [Effects of the Invention]

[0006] According to the present invention, editing of the causal relationship graph can be reflected in the order information. [Brief explanation of the drawings]

[0007] [Figure 1] Functional configuration diagram of the improvement system according to the first embodiment [Figure 2] FIG. 10 is a diagram showing an example of parameter information. [Figure 3] A diagram visually illustrating an example of order information. [Figure 4] FIG. 10 is a diagram showing an example of a display on an interface unit in the first embodiment; [Figure 5] An example of a causal graph [Figure 6] FIG. 10 is a diagram showing an example of a second correlation; [Figure 7] A diagram showing how a user interacts with a causal graph. [Figure 8] FIG. 10 is a diagram showing an example of a first correlation; [Figure 9] A visual representation of the processing of the order calculation unit [Figure 10] A visual representation of the processing of the order calculation unit [Figure 11] A visual representation of the processing of the order calculation unit [Figure 12] Flowchart showing the processing of the order calculation unit [Figure 13] Flowchart showing the processing of the order calculation unit [Figure 14] Functional configuration diagram of the improvement system according to the second embodiment [Figure 15] FIG. 10 is a diagram showing an example of a display on an interface unit in the second embodiment. [Figure 16] FIG. 10 is a diagram showing an example of layout information. [Figure 17] FIG. 13 is a diagram showing an example of a display on an interface unit in the third embodiment. [Figure 18] FIG. 13 is a diagram showing an example of a display on an interface unit in the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] -First embodiment- A first embodiment of the improvement system will be described below with reference to FIGS.

[0009] FIG. 1 is a functional configuration diagram of an improvement system 10. The improvement system 10 includes a first correlation calculation unit 101, a second correlation calculation unit 107, a graph generation unit 103, an order calculation unit 109, and an interface unit 110. The improvement system 10 includes one or more arithmetic units. The arithmetic unit includes, for example, a CPU (central processing unit), a ROM (read-only memory), and a RAM (readable and writable memory). The CPU loads programs stored in the ROM into the RAM and executes them to perform various calculations. Instead of a combination of a CPU, ROM, and RAM, the arithmetic unit may be realized by a field programmable gate array (FPGA), a rewritable logic circuit, or an application specific integrated circuit (ASIC). Furthermore, instead of a combination of a CPU, ROM, and RAM, the arithmetic unit may be realized by a different combination of configurations, for example, a combination of a CPU, ROM, RAM, and FPGA.

[0010] The first correlation calculation unit 101, the second correlation calculation unit 107, the graph generation unit 103, and the order calculation unit 109 are realized, for example, by a combination of the CPU, ROM, and RAM described above. The interface unit 110 is realized, for example, by a combination of a liquid crystal display and a pointing device, or by a liquid crystal display.

[0011] The improvement system 10 starts operation by reading parameter information 100 and order information 106. The parameter information 100 includes values ​​of evaluation indexes, specifically, KPIs (Key Performance Indicators), of multiple types of parameters. The order information 106 is information on the priority of all parameters included in the parameter information 100.

[0012] The first correlation calculation unit 101 reads the parameter information 100 and generates a first correlation 102. The graph generation unit 103 reads the first correlation 102 and the order information 106 and generates a causal relationship graph 104. The second correlation calculation unit 107 reads the first correlation 102 and the order information 106 and generates a second correlation 108. Note that the second correlation 108 will also be referred to as an "improvement effect" 108 below. A user 105 using the improvement system 10 can view and edit the causal relationship graph 104 and the second correlation 108 via the interface unit 110. When the user 105 edits the causal relationship graph 104, the order calculation unit 109 receives a notification and updates the order information 106. When the order information 106 is updated, the second correlation calculation unit 107 reads the first correlation 102 and the order information 106 again and generates the second correlation 108 again.

[0013] FIG. 2 is a diagram showing an example of parameter information 100. The parameter information 100 includes values ​​of evaluation indices for each perspective. A perspective is a common attribute for grouping events, such as a project, a product, a manufacturing location of the product, or a manufacturer. Examples of evaluation indices include a manufacturing process, a delivery date, labor hours, cost, and a delivery delay. In the example shown in FIG. 4, the perspective is described as "project." That is, the parameter information 100 shown in FIG. 2 includes values ​​of evaluation indices such as evaluation index A, evaluation index B, evaluation index C, and so on for each project such as project 1, project 2, project 3, and so on.

[0014] FIG. 3 is a diagram visually illustrating an example of the order information 106. The order information 106 includes information regarding the order of the evaluation indexes included in the parameter information 100. The order information 106 can be said to indicate the priority of the evaluation indexes, and in the case where the evaluation indexes are business processes, it is also the execution order of the business processes. The order of the evaluation indexes can be expressed using various methods, but FIG. 3 visually illustrates it for the sake of explanation. In the example shown in FIG. 3, the order is lower toward the left and higher toward the right. In addition, in FIG. 3, the up and down directions indicate the same order. For example, FIG. 3 shows that evaluation index D is ranked higher than evaluation index A, and that evaluation index A and evaluation index B have the same rank.

[0015] 4 is a diagram showing the display of the interface unit 110, specifically, the causal relationship graph 104 and the second correlation 108. In FIG. 4, the causal relationship graph 104 is displayed in the left half of the interface unit 110, and the second correlation 108 is displayed in the right half. The user 105 can operate the causal relationship graph 104 and the second correlation 108 displayed in the interface unit 110. The operation will be described later. The causal relationship graph 104 and the second correlation 108 will each be described with reference to another drawing.

[0016] FIG. 5 is a diagram illustrating an example of a causal relationship graph 104. The causal relationship graph 104 includes multiple points 702 representing evaluation indexes and directed lines 704 between the evaluation indexes. The causal relationship graph 104 can also be said to have evaluation indexes as nodes, with links connecting the nodes indicating the correlation between the evaluation indexes. In the causal relationship graph 104, nodes with strong inter-difference relationships, i.e., evaluation indexes, are connected by links, i.e., directed lines 704. A label 701 indicating the name of the evaluation index is displayed near the point 702 representing the evaluation index. An absolute value 703 of the correlation coefficient between the evaluation indexes is displayed near the directed line 704 connecting the evaluation indexes. If the correlation coefficient is negative, the directed line 704 between the evaluation indexes is displayed as a dashed line, as indicated by the symbol 705. In the example illustrated in FIG. 5, only nodes whose absolute value of the correlation coefficient is equal to or greater than 0.70, a threshold value defined within the system, are displayed.

[0017] For example, arrows extend from evaluation index A and evaluation index B toward evaluation index D, and a dashed arrow extends from evaluation index D to evaluation index G. This indicates that as the values ​​of evaluation index A and evaluation index B increase, the value of evaluation index D increases, and further, as the value of evaluation index D increases, the value of evaluation index G decreases.

[0018] When the user 105 operates the causal graph 104 displayed on the interface unit 110 and clicks a point representing an evaluation index with the mouse pointer 706, the interface unit 110 changes the color of the point to indicate that it has been selected, as indicated by reference numeral 707. Furthermore, important evaluation indexes defined within the system may be displayed in a color that can always be distinguished from other points, as indicated by reference numeral 708.

[0019] FIG. 6 is a diagram showing an example of the second correlation 108. The second correlation 108 includes one or more individual improvement effects 801. The individual improvement effect 801 consists of a sentence such as, "When 'E1' increases by 'N1', 'E2' is expected to increase (decrease) by 'N2'." In this sentence, 'E1' and 'E2' are the names of evaluation indicators, and 'N1' and 'N2' are numerical values. In this case, the part "When 'E1' increases by 'N1'" is called an improvement measure phrase, and the part "'E2' will increase (decrease) by 'N2'" is called an improvement effect phrase.

[0020] When the user 105 clicks a point 707 representing one of the evaluation indicators using the mouse pointer 706 in the causal relationship graph 104 displayed on the interface unit 110, the interface unit 110 performs the following process. That is, the interface unit 110 sets the point clicked by the user 105 as an improvement measure phrase and displays an individual improvement effect 801 on the interface unit 110, with the end point connected by a directed line from the clicked point as an improvement effect phrase. Furthermore, the interface unit 110 also displays an individual improvement effect 804, with the clicked point as an improvement measure phrase and with the key evaluation indicator defined within the system as displayed by reference numeral 708 as an improvement effect phrase.

[0021] The interface unit 110 may further accept editing by the user 105 of the numerical value of the measure numerical value 802 in the individual improvement effect 801. The interface unit 110 immediately updates the improvement effect numerical value 803 in response to the user's editing of the measure numerical value 802. For example, when the measure numerical value 802 is set to a value that is twice the current value, the interface unit 110 sets the improvement effect numerical value 803 to a value that is twice the current value. In other words, the interface unit 110 multiplies the quantity in the improvement effect 801 by a predetermined multiplier in response to the operation of the user 105 and displays the result.

[0022] FIG. 7 is a diagram showing operations on the causal relationship graph 104 by the user 105. FIG. 7 is roughly divided into four parts, starting with the first diagram 310 at the top left, passing through the second diagram 320 at the top right and the third diagram 330 at the bottom left, and ending in the fourth diagram 340 at the bottom right. Based on the operations of the user 105, the interface unit 110 rewrites the displayed causal relationship graph 104 as follows. An example of deleting and adding directed lines will be described below.

[0023] The user 105 moves the mouse pointer 311 to the interface unit 110, which is displaying the first diagram 310, and clicks on the directed line 312 to be deleted. The interface unit 110 then erases the clicked directed line 312, as shown in the second diagram 320. Next, the user 105 clicks on the start point 322 of the directed line to be added with the mouse pointer 321, and moves the mouse pointer 331 towards the end point 334 of the directed line to be added, and the interface unit 110 performs the following process.

[0024] That is, while the user 105 is moving the mouse pointer 321, the interface unit 110 displays a directed line 332 from a starting point 333 toward the mouse pointer 331, as shown in 330 of Figure 3. When the mouse pointer 331 reaches an end point 334, the interface unit 110 adds and displays a directed line 343 from a starting point 341 to an end point 342, as shown in 340 of Figure 4.

[0025] FIG. 8 is a diagram showing an example of the first correlation 102. The first correlation 102 shows the correlation between evaluation indexes. Each evaluation index is entered in a column 601 and a row 602. A cell 603 enters the correlation coefficient between the evaluation index of the column and the evaluation index of the row. Note that no value is entered when the evaluation index of the column and the evaluation index of the row indicate the same evaluation index, as in cell 604. The first correlation calculation unit 101 reads the parameter information 100, calculates the correlation coefficient between pairs of evaluation indexes for all cases, and outputs it as the first correlation 102. Note that this correlation is calculated by linear approximation, assuming that the correlation has linear characteristics.

[0026] The operation of the graph generation unit 103 will be described. The graph generation unit 103 first reads the first correlation 102 and the order information 106. Next, for cells whose absolute values ​​exceed a threshold defined within the system, the graph generation unit 103 extracts a pair of the evaluation index for the column and the evaluation index for the cell from the first correlation 102. The graph generation unit 103 then reads the ranking of each evaluation index in the pair from the order information 106. The graph generation unit 103 then generates a directed line from the lowest-ranked evaluation index to the highest-ranked evaluation index, and writes the absolute value of the correlation coefficient near the directed line. However, if the correlation coefficient of the pair is positive, the directed line is a solid line, and if it is negative, the directed line is a dashed line. In this way, a causal relationship graph 104 is generated and displayed on the interface unit 110 of the improvement system 10.

[0027] The operation of the second correlation calculation unit 107 will be described. The second correlation calculation unit 107 first reads the first correlation 102 and the order information 106. Next, for each cell in the first correlation 102 whose absolute value exceeds the threshold, the second correlation calculation unit 107 extracts a pair of the evaluation index for the column and the evaluation index for the cell. The second correlation calculation unit 107 then reads the order of the pair of evaluation indexes from the order information 106. The second correlation calculation unit 107 then uses the lower-ranked evaluation index as the explanatory variable and the higher-ranked evaluation index as the dependent variable, and calculates a linear equation for predicting the dependent variable from the explanatory variables using simple regression analysis. Finally, the second correlation calculation unit 107 converts the linear equation into text such as, "When 'E1' increases by 'N1', 'E2' is expected to increase (decrease) by 'N2'."

[0028] In this case, "E1" contains the name of the evaluation index with the lower ranking, and "E2" contains the name of the evaluation index with the higher ranking. "N1" is set to 1 as the initial value. "N2" contains the absolute value of the slope of the linear equation. If the slope of the linear equation is positive, it is considered to be "increase," and if the slope of the linear equation is negative, it is considered to be "decrease." The second correlation calculation unit 107 performs these processes on all evaluation index pairs of cells whose absolute values ​​exceed the aforementioned threshold, and saves the results as second correlation 108.

[0029] (Processing of the order calculation unit 109) 9 to 11 are diagrams visually illustrating the processing of the order calculation unit 109. When the user 105 changes the causal relationship graph 104 using the interface unit 110, the order calculation unit 109 modifies the order information 106 as follows. The order calculation unit 109 saves in advance the causal relationship graph 104 before editing by the user 105 (hereinafter referred to as "causal relationship graph A" 900). When the user 105 edits the causal relationship graph 104, the order calculation unit 109 receives a notification from the interface unit 110 that a change has been made, and reads the edited causal relationship graph 104 (hereinafter referred to as "causal relationship graph B" 901).

[0030] Then, the order calculation unit 109 performs a difference acquisition process 902 between the causal relationship graph A 900 and the causal relationship graph B 901, and extracts a difference evaluation index 903, which is the difference between the two. At this time, if the causal relationship graph A 900 is the first diagram 310 in Fig. 7 and the causal relationship graph B 901 is the fourth diagram 340 in Fig. 7, the difference evaluation index 903 will be as shown in Fig. 9.

[0031] Next, the order calculation unit 109 reads the pre-change order information 106 (hereinafter referred to as "pre-change order information" 904). The order calculation unit 109 then deletes all evaluation indexes included in the difference evaluation index 903 from the pre-change order information 904, and performs column addition processing 905 to add columns equal to the number of difference evaluation indexes 903 to each column where an evaluation index exists, thereby generating pre-addition order information 906. For example, if the user 105 changes the directed graph from evaluation index A to evaluation index D in the opposite direction as in the example of FIG. 4, evaluation index A and evaluation index D are deleted from the pre-change order information 904 as shown in FIG. 10(a), and two columns are added to each column where an evaluation index exists, resulting in the state shown in FIG. 10(b). Note that P1 to P14 in FIG. 10(b) are shown for later explanation.

[0032] Next, the order calculation unit 109 adds the difference evaluation index 903 to the pre-addition order information 906, and performs temporary addition process 907 to obtain temporary order information 908. In temporary addition process 907, it is essential to reflect the order represented in the difference evaluation index 903, i.e., the order of evaluation index A being higher than evaluation index D, but since the relationship of the ranking with other evaluation indexes is not immediately clear, one of the expected rankings is selected. The temporary addition process 907 for generating the temporary order information 908 will be specifically described with reference to FIGS. 10 and 11.

[0033] 10(b), since evaluation index A has a higher order than evaluation index D as described above, evaluation index A is placed to the right of evaluation index D. For P1 to P14 where evaluation index A and evaluation index D can be placed, the only constraint is their relative positions, so the number of possible placement combinations is enormous.

[0034] For example, evaluation index D can be placed at any of P1 to P14 except for P14, which is on the far right, and when evaluation index D is placed at P6, evaluation index A can be placed at any of P7 to P14. The order calculation unit 109 selects any one combination from this vast number of combinations, and places evaluation index A and evaluation index D to generate tentative order information 908. The tentative order information 908 shown in Figure 11(a) shows an example in which evaluation index D is placed at P2 in the pre-addition order information 906 and evaluation index A is placed at P12 in the pre-addition order information 906. Returning to Figure 9, the explanation will continue.

[0035] Next, the order calculation unit 109 creates a temporary causal relationship graph 104 (hereinafter referred to as "causal relationship graph C" 910) using the most recently created tentative order information 908 and the first correlation 102. Note that the order calculation unit 109 may cause the graph generation unit 103 to generate the causal relationship graph C 910. The order calculation unit 109 then performs a comparison process 911 between the causal relationship graph B 901 edited by the user and the causal relationship graph C 910. The order calculation unit 109 then performs a match determination 912 to determine the result of the comparison process 911, and if it determines that the two do not match, it returns to the tentative addition process 907 and selects another combination. If it determines that the two match, the order calculation unit 109 performs an update process 913 to overwrite the order information 106 with the latest tentative order information 908. This concludes the explanation of FIG. 9.

[0036] (flowchart) 12 and 13 are flowcharts showing the processing of the order calculation unit 109, and summarize the processing described with reference to Figures 9 to 11. The order calculation unit 109 is responsible for executing each step described below.

[0037] In step S1001, the order calculation unit 109 acquires a causal relationship graph A900, which is the pre-edited causal relationship graph 104. In the following step S1002, the order calculation unit 109 acquires a causal relationship graph B901, which is the edited causal relationship graph 104. In the following step S103, the order calculation unit 109 acquires all evaluation indexes in the causal relationship graph B901 that have a different line connection from the causal relationship graph A900, i.e., differential evaluation indexes 903.

[0038] In the following step S1004, the order calculation unit 109 calculates the number of evaluation indexes in the difference evaluation index 903 and stores it in variable d. In the following step S1005, the order calculation unit 109 acquires pre-change order information 904, i.e., the current order information 106. In the following step S1006, the order calculation unit 109 deletes the evaluation indexes that exist in the difference evaluation index from the pre-change order information 904, and creates pre-addition order information 906. In the following step S1007, the order calculation unit 109 adds d columns to the right of each column that has an evaluation index in the pre-change order information 904. Thereafter, the order calculation unit 109 proceeds to step S1008 in the next drawing via circled A.

[0039] 13, in step S1008, the order calculation unit 109 calculates all combinations of candidates for positions in the pre-change order information 904 where the difference evaluation index 903 is to be added, and the set of these combinations is referred to as N. Hereinafter, the order calculation unit 109 repeats the processes of steps S1009 to S1003 by selecting one pair at a time from set N. In step S1009, the order calculation unit 109 selects one unselected pair from set N, adds an evaluation index to the pre-change order information 904, and temporarily saves it as tentative order information 908. In the following step S1010, the order calculation unit 109 generates a causality graph C910 from the tentative order information 908 and the first correlation 102.

[0040] In the following step S1011, the order calculation unit 109 compares the causal relationship graph B901 with the causal relationship graph C910. In the following step S1012, if the order calculation unit 109 determines that the causal relationship graph B901 and the causal relationship graph C910 match as a result of the comparison, the process proceeds to step S1015, and if it determines that they do not match, the process proceeds to step S1013.

[0041] In step S1013, the order calculation unit 109 discards the tentative order information 908, and in the following step S1014, the process returns to step S1019. In step S1015, the order calculation unit 109 saves the tentative order information 908 as the corrected order information 106, and ends the processing shown in FIGS.

[0042] According to the first embodiment described above, the following advantageous effects can be obtained. (1) The improvement system 10 includes a first correlation calculation unit 101 that reads parameter information 100, which is information about a plurality of evaluation indicators in a business process, and calculates a first correlation 102 that expresses the correlation between the evaluation indicators as a correlation coefficient; a graph generation unit 103 that creates a causal relationship graph 104 in which the evaluation indicators are used as nodes and the correlation between the evaluation indicators is indicated by links, based on the first correlation 102 and order information 106, which is information about the priorities of the plurality of evaluation indicators; an interface unit 110 that presents the causal relationship graph 104 to a user 105 and accepts changes to the causal relationship graph 104 made by the user 105; and an order calculation unit 109 that updates the order information 106 based on the changes made to the causal relationship graph 104 by the user 105, enabling the graph generation unit 103 to generate the causal relationship graph 104 after the changes made by the user 105. Therefore, edits made by the user 105 to the causal relationship graph 104 can be reflected in the order information 106, and the user 105 can rewrite the order information 106 without performing complex operations.

[0043] (2) The improvement system 10 includes a second correlation calculation unit 107 that generates a second correlation 108 that expresses in words the correlation between the quantities of the evaluation indexes, based on the parameter information 100 and the order information 106. When the order calculation unit 109 updates the order information 106, the second correlation calculation unit 107 generates the second correlation 108 based on the updated order information 106 and the parameter information 100. Therefore, the improvement system 10 can express in words the correlation between the evaluation indexes.

[0044] (3) The graph generating unit 103 displays evaluation indices whose correlation coefficients are equal to or greater than a predetermined threshold, for example, 0.70, on the causal relationship graph 104, as shown in Fig. 5. Therefore, only evaluation indices with strong correlations can be presented to the user 105.

[0045] (4) The first correlation calculation unit 101 calculates the correlation coefficient by linear approximation, which allows the correlation coefficient to be calculated by simple calculation.

[0046] 9 , the order calculation unit 109 extracts evaluation indexes related to changes to the causal relationship graph by the user 105 as difference evaluation indexes 903, creates multiple pieces of tentative order information 908 by changing the order of the difference evaluation indexes 903 in the order information 106, and sets the tentative order information 908 in which the causal relationship graph C910 created by the graph generation unit 103 based on the tentative order information 908 and the parameter information 100 matches the causal relationship graph B901 after the change by the user 105 as the updated order information 106. Therefore, the order calculation unit 109 can calculate the order information 106 that matches the changes made by the user 105 without performing complex calculations.

[0047] (6) For evaluation indices whose correlation coefficients are equal to or greater than a predetermined threshold, the graph generation unit 103 creates a directed line as a link, starting from a higher-level evaluation indices in the order information 106 and ending at a lower-level evaluation indices in the order information. This allows the correlation between evaluation indices to be visually presented to the user 105.

[0048] (7) The interface unit 110 accepts an operation by the user 105 to change the links in the causal relationship graph 104. Therefore, the improvement system 10 can reflect the link changes made by the user 105 in the order information 106.

[0049] (8) The interface unit 110 presents the second correlation related to the selected node to the user based on the node selection by the user 105. This allows the user 105 to be provided with information about evaluation indicators that interest him / her.

[0050] (9) The interface unit 110 multiplies the quantity in the second correlation by a predetermined magnification and displays it based on the operation of the user 105. Therefore, the second correlation can be presented to the user 105 in an easy-to-understand manner. (10) The second correlation calculation unit 107 calculates the regression coefficient between the evaluation indexes as the second correlation. (11) The evaluation index includes at least one of the man-hours and the cost.

[0051] (Variation 1) In the above-described embodiment, a liquid crystal display and a pointing device are exemplified as the configuration of interface unit 110, and a configuration is shown in which information is directly presented to user 105 and information is directly acquired from user 105. However, interface unit 110 may be configured to indirectly exchange information with user 105, and for example, interface unit 110 may be configured to input and output information to and from a device held by user 105.

[0052] --Second embodiment-- A second embodiment of the improvement system will be described with reference to Figs. 14 to 16. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that it accepts layout changes made by the user.

[0053] In this embodiment, layout modifications to the causal relationship graph 104 are accepted from the user 105. The layout modifications here refer to modifications to the X and Y coordinates of each node in the causal relationship graph 104, and the relationships of the directed lines between the nodes remain unchanged before and after the layout modifications.

[0054] Fig. 14 is a functional configuration diagram of the improvement system 10A in the second embodiment. In Fig. 14, layout information 1100 is added compared to Fig. 1, and input and saving are performed based on the operation of the interface unit 110 by the user 105. That is, the interface unit 110 reads the existing layout information 1100 and reflects it in the display, and when the user 105 edits the layout, the changed layout information is saved as the layout information 1100. In addition, the interface unit 110 may automatically read the layout information 1100 when the improvement system 10A is started up.

[0055] Fig. 15 is a diagram showing the display of the interface unit 110 in the second embodiment. Compared to Fig. 4, Fig. 15 has a layout operation unit 1200 added to the lower right of the figure, and a "view / edit switch" button 1203 and an "edit order" button 1204 added to the upper part of the figure. The layout operation unit 1200 includes a layout pattern 1201 that can switch between multiple layouts, and a layout save button 1202 for saving processing.

[0056] When the user 105 selects one layout pattern from layout patterns 1201 in the layout operation unit 1200, the interface unit 110 modifies the layout of the causal relationship graph 104 based on the layout information 1100. When the user 105 presses a "switch between viewing and editing" button 1203, the user 105 becomes able to drag and drop each node of the causal relationship graph 104, thereby modifying the layout. When the user 105 has finished modifying the layout of the causal relationship graph 104 and wishes to edit the order relationship, the user 105 selects an "edit order" button 1204. Furthermore, when the user 105 presses a save layout button 1202, the interface unit 110 saves the currently displayed layout in the layout information 1100.

[0057] 16 is a diagram showing an example of layout information 1150. Layout information 1150 has evaluation index names in column 1301, X coordinate information in column 1302, and Y coordinate information in column 1303. Layout information 1150 is rewritten by interface unit 110 based on operations by user 105.

[0058] According to the second embodiment described above, the position of each evaluation index on the causal relationship graph 104 can be freely changed so that the user 105 can easily view it.

[0059] -Third embodiment- A third embodiment of the improvement system will be described with reference to Fig. 17. In the following description, the same components as those in the first embodiment are given the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that detailed information about nodes is displayed in the interface section.

[0060] FIG. 17 is a diagram showing an example of a display of the interface unit 110 in the third embodiment. In FIG. 17, a detailed information display unit 1400 is added to FIG. 4. When the user 105 selects a node of interest in the causal relationship graph 104 displayed on the interface unit 110, the interface unit 110 performs the following operation. Specifically, the interface unit 110 displays the improvement effect 108 related to the selected node in the upper right corner of the figure, and further displays detailed information about each evaluation index included in the improvement effect 108 in the detailed information display unit 1400 in the lower right corner of the figure. The detailed information includes statistical data such as correlation coefficients, number of data, average values, and medians. The evaluation indexes of the node selected by the user 105 may be displayed in a different manner from the others, as indicated by reference numeral 1402.

[0061] According to the third embodiment described above, detailed information on the evaluation index that the user 105 has focused on can be displayed on the interface unit 110.

[0062] --Fourth embodiment-- A fourth embodiment of the improvement system will be described with reference to Fig. 18. In the following description, the same components as those in the first embodiment are given the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that each node can be individually set to be hidden in the interface unit.

[0063] Fig. 18 is a diagram showing an example of the display of the interface unit 110 in the fourth embodiment. Fig. 18 differs from Fig. 4 in that a node display setting unit 1500 is added. The node display setting unit 1500 displays "ON" or "OFF" for each evaluation index. "ON" indicates that the index is to be displayed on the interface unit 110, and "OFF" indicates that the index is not to be displayed on the interface unit 110.

[0064] For example, each time the user 105 selects an item in the node display setting unit 1500, the setting is switched between "ON" and "OFF." When an evaluation index is set to "OFF," the interface unit 110 hides not only the evaluation index but also the directed lines connected to the evaluation index. When an evaluation index is set to "ON," the interface unit 110 displays not only the evaluation index but also the directed lines connected to the evaluation index.

[0065] According to the above-described fourth embodiment, the following advantageous effects can be obtained. (12) The interface unit 110 hides one or more evaluation indices based on the operation of the user 105. Therefore, evaluation indices that are unnecessary for the user 105 are not displayed, and only information on the evaluation indices that are necessary can be provided.

[0066] In each of the above-described embodiments and modifications, the functional block configurations are merely examples. Some functional configurations shown as separate functional blocks may be configured as an integrated unit, or a configuration shown in a single functional block diagram may be divided into two or more functions. Furthermore, some of the functions of each functional block may be provided by other functional blocks.

[0067] In the above-described embodiments and modifications, the program is stored in a ROM (not shown), but the program may be stored in a non-volatile storage device included in the improvement system 10. Furthermore, the improvement system 10 may be provided with an input / output interface (not shown), and the program may be loaded from another device via a medium available to the input / output interface when necessary. Here, the medium refers to, for example, a storage medium detachable from the input / output interface, or a communication medium, i.e., a wired, wireless, or optical network, or a carrier wave or digital signal propagating through the network. Furthermore, some or all of the functions realized by the program may be realized by a hardware circuit or FPGA.

[0068] The above-described embodiments and modifications may be combined with each other. Although various embodiments and modifications have been described above, the present invention is not limited to these. Other embodiments conceivable within the scope of the technical concept of the present invention are also included within the scope of the present invention. [Explanation of symbols]

[0069] 10...Improvement System 100...Parameter information 101...First correlation calculation unit 102...First correlation 103...Graph generation unit 104...Causal relationship graph 105...User 106…Order information 107...Second correlation calculation unit 108...Second correlation, improvement effect 109...Order calculation part 110...Interface section

Claims

1. a first correlation calculation unit that reads parameter information that is information about a plurality of evaluation indexes in a business process and calculates a first correlation that expresses the correlation between the evaluation indexes as a correlation coefficient; a graph generation unit that generates a causal relationship graph in which the evaluation indexes are nodes and correlations between the evaluation indexes are indicated by links, based on the first correlation and order information that is information regarding the priority of the evaluation indexes; an interface unit that presents the causal relationship graph to a user and accepts changes to the causal relationship graph made by the user; an order calculation unit that extracts the evaluation indexes related to the changes made to the causal relationship graph by the user as difference evaluation indexes, creates multiple sets of tentative order information by changing the order of the difference evaluation indexes in the order information, and sets the tentative order information that matches the temporary causal relationship graph created by the graph generation unit based on the tentative order information and the parameter information as the updated order information.

2. 10. The remediation system of claim 1, a second correlation calculation unit that, when the order information is updated by the order calculation unit, performs a simple regression analysis using the lower-order evaluation index in the order information as an explanatory variable and the higher-order evaluation index in the order information as a response variable, and calculates, as an improvement effect, a value based on a slope and a sign of a linear term in a regression equation of the simple regression analysis; The interface unit presents the updated order information and the improvement effect calculated by the second correlation calculation unit.

3. 3. The remediation system of claim 2, The interface unit presents the improvement effect associated with the selected node to the user based on the node selection by the user.

4. 3. The remediation system of claim 2, The interface unit displays the quantity of the improvement effect multiplied by a predetermined multiplier based on the user's operation.

5. 10. The remediation system of claim 1, An improvement system in which, when creating the provisional order information, the order calculation unit reflects the order between the evaluation indexes included in the difference evaluation indexes the order between the evaluation indexes deleted from the pre-change order information.

6. 1. An improved method implemented by one or more computers, comprising: Reads parameter information, which is information on multiple evaluation indicators in a business process, a first correlation calculation step of calculating a first correlation that expresses the correlation between the evaluation indexes as a correlation coefficient; a graph generation step of generating a causal relationship graph in which the evaluation indexes are nodes and correlations between the evaluation indexes are indicated by links, based on the first correlation and order information which is information regarding the priority of the plurality of evaluation indexes; an input step of presenting the causality graph to a user and accepting changes to the causality graph by the user; an order calculation step of extracting the evaluation indexes related to the changes made to the causal relationship graph by the user as difference evaluation indexes, creating multiple sets of tentative order information by changing the order of the difference evaluation indexes in the order information, and setting the tentative order information, which is a temporary causal relationship graph created by the graph generation step based on the tentative order information and the parameter information, to be the updated order information and which coincides with the causal relationship graph after the changes made by the user.

Images