Causal relationship estimation system and production support system

Abstract

To provide a causal relationship estimation system and a production support system for mechanically selecting an appropriate algorithm for data to be analyzed.SOLUTION: A causal relationship estimation system 2 includes: a storage apparatus 2a which stores data DA to be analyzed and a plurality of different causal inference algorithms CA; and a computing apparatus 2b which estimates a causal relationship included in the data DA using the causal inference algorithms CA. The computing apparatus 2b performs regression analysis using one determination algorithm CAa selected from among the different causal inference algorithms CA for the data DA to calculate a residual R, calculates a p value P representing independence between the residual R and an explanatory variable corresponding to the residual R, determines, using the p value P, a causal inference algorithm CA to be adopted from among the different causal inference algorithms CA, and estimates a causal relationship based on the adopted causal inference algorithm CA.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a causal relationship estimation system and a production support system. [Background technology]

[0002] Conventionally, Patent Document 1 discloses a technology for estimating causal relationships within data to be analyzed using structural equation modeling means and generating a nonlinear regression model based on the estimation results. Several algorithms have been proposed for structural equation modeling. By applying an appropriate algorithm to various types of data to be analyzed, it is possible to infer causal relationships in the data to be analyzed. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2006-099482 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there is no established method for automatically determining which algorithm is best for the data being analyzed. Therefore, in the past, algorithms were selected manually, which required a lot of effort.

[0005] The present invention has been made in view of the above-mentioned problems, and aims to provide a causal relationship estimation system and a production support system that can mechanically select an appropriate algorithm for data to be analyzed. [Means for solving the problem]

[0006] One aspect of the present invention is a storage device for storing data to be analyzed and a plurality of different causal inference algorithms; a calculation device that estimates a causal relationship included in the analysis target data using one causal inference algorithm selected from the plurality of different causal inference algorithms, The computing device performing a regression analysis on the analysis target data using one determination algorithm selected from the plurality of different causal inference algorithms to calculate a residual; calculating an independence index representing the independence between the residual and an explanatory variable corresponding to the residual; determining which causal inference algorithm to adopt from among the plurality of different causal inference algorithms using the independence index; The causal relationship inference system infers causal relationships based on the adopted causal inference algorithm.

[0007] Another aspect of the present invention is The above causal relationship estimation system, a server or virtual server that stores a knowledge database that indicates the relationship between information or processing specifications related to a workpiece and a processing result that is the result of performing a predetermined processing on the workpiece; and a communication device that acquires data to be analyzed from the machine tool or the production system by communication. [Effects of the Invention]

[0008] According to one and other aspects of the present invention, a residual is calculated using a judgment algorithm, an independence index representing the independence between this residual and the explanatory variable corresponding to the residual is calculated, and by using this independence index, an appropriate causal inference algorithm can be mechanically selected for the data to be analyzed. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a causal relationship estimation system and a production support system according to a first embodiment. [Figure 2]FIG. 1 is a diagram showing a network diagram representing a knowledge model according to the first embodiment. [Figure 3] 1 is a block diagram showing a storage device according to a first embodiment. [Figure 4] 4 is a flowchart showing the operation of ANM processing according to the first embodiment. [Figure 5] 1 is a flowchart showing the operation of RECI processing according to the first embodiment. [Figure 6] 3 is a main flow of the operation of the causal relationship estimation system according to the first embodiment. [Figure 7] 10 is a part of a flowchart showing the operation of an algorithm determination process according to the first embodiment. [Figure 8] 10 is a part of a flowchart showing the operation of an algorithm determination process according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] (Embodiment 1) 1. Configuration of the causal relationship estimation system and production support system A production support system 1 according to this embodiment will be described with reference to Fig. 1. The production support system 1 includes a causal relationship estimation system 2, a server 3, a machine tool 4, and a production system 5. The causal relationship estimation system 2, the server 3, the machine tool 4, and the production system 5 are connected to a network 6 and configured to be able to communicate with each other. However, one of the machine tool 4 and the production system 5 may be omitted. Also, there may be two or more machine tools 4, and there may also be two or more production systems 5.

[0011] The causal relationship estimation system 2 includes a storage device 2a, a calculation device 2b, an input device 2c, an output device 2d, a communication device 2e, and an update device 2f.

[0012] The server 3 stores a knowledge database DB, but the server 3 may be a virtual server on the Internet or an intranet.

[0013] The machine tool 4 performs predetermined processing on a workpiece (not shown). The machine tool 4 is not particularly limited, and may be, for example, a lathe, a machining center, a milling machine, a gear processing machine, or a boring machine.

[0014] The production system 5 comprises a management device 5a, a data processing device 5b, and a machine tool 5c. However, the configuration of the production system 5 is not limited to the above and can be selected as appropriate. The management device 5a stores, for example, a production plan. The management device 5a may also have a function for generating a production plan. The production plan includes management data such as the production deadline (delivery date) for each workpiece, process information for the workpiece, and an execution schedule for each process. The data processing device 5b determines command information for the operation of the machine tool 4 based on the production plan in the management device 5a, and outputs the command information to the machine tool 5c. The machine tool 5c in the production system 5 is similar to the machine tool 4 described above, so a redundant description will be omitted.

[0015] The server 3 stores a knowledge database DB. Generally, in the field of machining, an operator determines machining conditions such as cutting speed and cutting depth per unit time, taking into consideration various information such as the material of the workpiece, the material of the tool, the quality of the workpiece, and the machining cycle time. In this case, the knowledge database DB is a model of the operator's thought process when the operator acquires various input information and determines the machining conditions.

[0016] In other words, in addition to the material of the workpiece, the material of the tool, the quality of the workpiece, the processing cycle time, the cutting speed, the depth of cut, etc., the knowledge database DB defines each of the industrial technology elements that appear in the worker's thought process as factors, and defines the relationships between the factors.

[0017] The knowledge database DB is used, for example, as follows: When an operator inputs necessary information about the input factors, such as the material of the workpiece, the material of the tool, and the machining requirements (machining accuracy, machining cycle time, etc.), information about the cutting speed, cutting depth, etc., as output factors is output.

[0018] A knowledge database DB is conceptually expressed in the form of a network. An example of a knowledge network diagram 10 that expresses a knowledge database DB as a network diagram will be described with reference to FIG. 2. In this example, a knowledge network diagram 10 relating to a knowledge database DB in the machining field will be used as an example. However, the knowledge network diagram 10 is not limited to FIG. 2.

[0019] As shown in FIG. 2, the knowledge network diagram 10 comprises a plurality of node graphics 11 and link graphics 12 connecting the node graphics 11 together. The node graphics 11 are represented by any graphic such as a box, a graphic including text, an icon, etc. The node graphics 11 represent factors in the knowledge database DB. The link graphics 12 are represented by straight lines, curves, elbow lines, etc. In this example, the link graphics 12 are represented by arrows to define the directionality of the relationships. The link graphics 12 represent the relationships connecting factors in the knowledge database DB. In the knowledge network diagram 10 shown in FIG. 2, all of the node graphics 11 are represented by boxes in which text can be written, and the link graphics 12 are represented by arrows.

[0020] Here, factors are defined by technical terms. Multiple factors may have a technical inclusive relationship (also called a hierarchical relationship, parent-child relationship, or master-slave relationship), or they may have a technical heterogeneous relationship. In other words, the relationships between factors are classified into the above two types.

[0021] For example, workpiece specifications have an inclusive relationship with workpiece thermal properties, workpiece hardness, workpiece elongation, etc. In other words, as factors having a technical inclusive relationship, workpiece specifications are considered to be superordinate conceptual factors, and workpiece thermal properties, workpiece hardness, workpiece elongation, etc. are considered to be subordinate conceptual factors. For example, factors having a technically heterogeneous relationship include workpiece thermal properties and required tool heat resistance, etc. In the following, the relationship between two factors having a technical inclusive relationship will be simply referred to as an inclusive relationship, and the relationship between two factors having a technically heterogeneous relationship will be simply referred to as a heterogeneous relationship.

[0022] The link graphics 12 are displayed in such a way that a first link graphic 12a representing an inclusive relationship and a second link graphic 12b representing a heterogeneous relationship are distinguished from each other. That is, the first link graphic 12a and the second link graphic 12b are displayed in different ways.

[0023] 2, the first link graphic 12a representing the inclusion relationship is represented by a frame line indicating the area of ​​the superordinate concept factor, and the subordinate concept factors are arranged within the frame line representing the first link graphic 12a. Note that the first link graphic 12a is not limited to a frame line.

[0024] 2, the second link graphic 12b representing the heterogeneous relationship is represented by a straight line, a broken line, or the like, connecting the node graphics 11 placed at any positions (up, down, left, right) apart. The second link graphic 12b is represented by an arrow line indicating the direction of the definition of the factors.

[0025] The following are examples of factors that are effective when the causal relationship estimation system 2 according to this embodiment is applied to the production system 5. The factors include the power value and torque of the spindle motor. Other factors include thermal displacement of the machine tool 4 and shape error due to thermal displacement. However, the factors that are effective in the causal relationship estimation system 2 according to this embodiment are not limited to those mentioned above.

[0026] 2. Configuration of Causal Relationship Inference System 2 The configuration of the causal relationship estimation system 2 will be described with reference to FIGS.

[0027] (1) Storage device 2a As shown in FIG. 3, the storage device 2a stores analysis target data DA, a causal inference algorithm CA, a residual R, a causal direction index CI, a p-value P, a causal direction value dir, and a threshold value TH.

[0028] The data DA to be analyzed includes machine setting conditions or machining conditions of the machine tool 4 or production system 5 that executes a predetermined process on the workpiece, and data related to the processing results of executing the predetermined process on the workpiece. The data DA to be analyzed includes two variables. One of the two variables is designated X and the other is designated Y. In this embodiment, a causal relationship is inferred between one of the two variables (X) as the cause and the other (Y) as the result, and also between the other variable (Y) as the cause and the one variable (X) as the result.

[0029] The causal inference algorithm CA is an algorithm for estimating causal relationships in structural equation modeling. The causal inference algorithm CA is not particularly limited and may be appropriately selected from, for example, the Addettive Noise Model (ANM), Regression Error Based Causal Inference (RECI), Linear Non-Gaussian Acyclic Model (LiNGAM), Information-Geometric Causal Inference (IGCI), Post Nonlinear Causal model (PNL), etc.

[0030] A predetermined assumption is set for data to which each causal inference algorithm CA can be applied. In other words, when a causal inference algorithm CA in which an assumption corresponding to data having predetermined properties is set is applied, highly accurate causal inference can be performed, but when a causal inference algorithm CA in which an assumption not corresponding to the properties of the data is set is applied, the accuracy of the causal inference may decrease.

[0031] ANM is based on assumptions such as that the data is three times differentiable and that the explanatory variables and the residual R are independent.

[0032] RECI is based on assumptions such as that the data is twice differentiable.

[0033] LiNGAM is based on assumptions such as that the data is linear and that explanatory variables and residual R are independent.

[0034] The IGCI is based on assumptions such as nonlinearity and the absence of noise in the data.

[0035] The residual R is the difference between the actual measured value and the theoretical value according to the model estimated by the causal inference algorithm CA.

[0036] The causal direction index CI is calculated by applying a plurality of different causal inference algorithms CA to the analysis target data DA. The causal direction index CI is not particularly limited and includes, for example, a score C calculated by using ANM and a mean square error MSE calculated by using RECI.

[0037] The p-value P is an example of an independence index that indicates the independence of the explanatory variables corresponding to the residual R. Note that the independence index is not limited to the p-value, and may be a statistic or the like.

[0038] The p-value P in this embodiment refers to the probability that a value more extreme than the observed value will be obtained when the null hypothesis is assumed to be correct in a statistical hypothesis test. The p-value P can be used as a criterion for determining whether the set hypothesis is correct in a statistical hypothesis test. A hypothesis test is a method for statistically determining whether the hypothesis you have set is correct. The p-value P in this embodiment is the p-value P where the null hypothesis is that the explanatory variables and the residual R are independent.

[0039] The causal direction value dir is data relating to information that, of two variables (X and Y) in the analysis target data DA, X is the cause and Y is the result (X → Y), information that X is the result and Y is the cause (Y → X), or information that it is impossible to estimate. The causal direction value dir can be set to, for example, "1" when X → Y, "-1" when Y → X, or "0" when it is impossible to predict, and any numerical value can be selected as appropriate.

[0040] The threshold value TH is used to determine the independence of the explanatory variables with respect to the residual error R by comparing with the independence index described above. The value of the threshold value TH is not particularly limited, and any value can be adopted. When the p-value P is used as the independence index, for example, 0.05 or 0.01 can be adopted.

[0041] (2) Arithmetic unit 2b The calculation device 2b shown in FIG. 1 calculates a residual R by performing a regression analysis on analysis target data DA using one determination algorithm CAa selected from a plurality of different causal inference algorithms CA. The calculation device 2b calculates an independence index that indicates the independence between the residual R and an explanatory variable for the residual R. The calculation device 2b uses the independence index to determine which causal inference algorithm CA to adopt from the plurality of different causal inference algorithms CA. The calculation device 2b estimates a causal relationship based on the adopted causal inference algorithm CA.

[0042] The arithmetic device 2b may be, for example, a central processing unit (CPU) or a graphics processing unit (GPU).

[0043] (3) Input device 2c The input device 2c shown in FIG. 1 is not particularly limited, and any input device 2c can be selected, such as a keyboard, a mouse, a touch panel, a joystick, or a voice input device.

[0044] (4) Output device 2d 1 is not particularly limited, and any output device 2d can be selected, such as a liquid crystal display, an audio output device, a tablet terminal, a printer, etc. The output device 2d outputs the causal relationship estimated by the causal relationship estimation system 2 and notifies the worker of it.

[0045] (5) Communication device 2e 1 acquires analysis target data DA through communication from the machine tool 4 or the production system 5. In addition, the communication device 2e transmits update information of the knowledge database DB to the server 3.

[0046] (6) Update device 2f When the causal relationship estimation system 2 updates the causal relationships related to the analysis target data DA, the update device 2f shown in FIG. 1 updates the corresponding portion of the knowledge database DB.

[0047] 3. Explanation of the causal inference algorithm CA As an explanation of the processing of the causal inference algorithm CA, an ANM processing (S1) using ANM and a RECI processing (S2) using RECI will be explained with reference to FIGS.

[0048] 4 shows a flowchart of the ANM process (S1). When the ANM process (S1) is executed, the calculation device 2b performs a regression analysis by applying ANM to the analysis target data DA (S10). The calculation device 2b calculates a residual R from the predicted value by the regression analysis and the actual measured value (S11).

[0049]

number

[0050]

number

[0051] Next, the calculation device 2b calculates a score C for both directions of the causal relationship (X→Y and Y→X) (S12). The score C is an example of a causal direction index CI.

[0052]

number

[0053]

number

[0054] Next, the calculation device 2b calculates C X→Y and C Y→X It is determined whether they are equal (S13).

[0055] C X→Y And C Y→X If and are equal (S13: Y), the calculation device 2b determines that it is impossible to infer a causal relationship by ANM, and assigns information that it is impossible to infer a causal relationship to the causal direction value dir (S14). For example, if the causal direction value dir ANM 0 is substituted for the causal direction value dir. ANM is the causal direction value dir calculated by ANM. Next, a causal relationship is estimated based on the estimated causal direction (S18). This completes the ANM process (S1).

[0056] On the other hand, C X→Y And C Y→X If and are not equal (S13:N), the calculation unit 2b calculates C X→Y But C Y→X The calculation device 2b determines whether the value is smaller than CX→Y But C Y→X If it is determined that the causal direction is smaller than (S15: Y), the variable X is inferred as the cause and the variable Y as the result, and the information that the causal direction is X → Y is substituted into the causal direction value dir (S16). For example, if the causal direction value dir ANM Then, 1 is substituted into . Next, the causal relationship is estimated based on the estimated causal direction (S18). This completes the ANM process (S1).

[0057] On the other hand, the calculation unit 2b X→Y But C Y→X If it is determined that the causal direction is not smaller than (S15: N), the variable Y is inferred as the cause and the variable X as the result, and the information that the causal direction is Y → X is substituted into the causal direction value dir (S17). For example, if the causal direction value dir ANM Then, the causal relationship is estimated based on the estimated causal direction (S18). This completes the ANM process (S1).

[0058] Next, a flowchart of the RECI process (S2) is shown in Fig. 5. When the RECI process (S2) is executed, the calculation device 2b performs regression analysis by applying RECI to the analysis target data DA (S20).

[0059] Next, the calculation device 2b calculates the mean square error MSE for both directions of the causality (S21). The mean square error MSE is an example of a causality direction index CI.

[0060]

number

[0061]

number

[0062] Next, the calculation device 2b calculates the MSE Y|X and MSE X|Y It is determined whether they are equal (S22).

[0063] MSE Y|X and MSE X|Y If RECI is equal to dir (S22: Y), the calculation device 2b determines that it is impossible to infer a causal relationship from RECI, and assigns information that it is impossible to infer a causal relationship to the causal direction value dir (S23). For example, if the causal direction value dir RECI 0 is substituted for the causal direction value dir. RECI is the causal direction value dir calculated by RECI. Next, a causal relationship is estimated based on the estimated causal direction (S27). This completes the RECI process (S2).

[0064] On the other hand, MSE Y|X and MSE X|Y If and are not equal (S22:N), the calculation device 2b calculates MSE Y|X But MSE X|Y The calculation unit 2b determines whether the MSE Y|X But MSE X|Y If it is determined that the value is smaller than dir (S24: Y), the causal direction is inferred to be that variable X is the cause and variable Y is the result, and the causal direction value dir RECI For example, 1 is substituted into the causal direction value dir. Next, the causal relationship is estimated based on the estimated causal direction (S27). This completes the RECI process (S2).

[0065] On the other hand, the calculation unit 2b performs MSE Y|X But MSE X|YIf it is determined that the variable Y is not smaller than 1 (S24:N), the causal direction is inferred to be the cause and the variable X is the result, and the causal direction value dir RECI The information that the causal direction is Y → X is substituted into the causal direction value dir (S26). For example, -1 is substituted into the causal direction value dir. Next, the causal relationship is estimated based on the estimated causal direction (S27). This completes the RECI process (S2).

[0066] 4. Operation of Causality Inference System 2 Next, the operation of the causal relationship estimation system 2 will be described with reference to Fig. 6 to Fig. 8. Fig. 6 shows a main flow of the causal relationship estimation system 2 according to this embodiment. When the causal relationship estimation system 2 is started, an algorithm determination process (S30) is executed. By executing the algorithm determination process (S30), one causal inference algorithm CA is selected from a plurality of different causal inference algorithms CA. In this embodiment, either ANM or RECI is selected.

[0067] If ANM is selected by the algorithm determination process (S30), ANM process (S1) is executed, and if RECI is selected, RECI process (S2) is executed. However, since the ANM process (S1) in FIG. 6 is the same as the ANM process (S1) in FIG. 4, a duplicated explanation will be omitted. Furthermore, since the RECI process (S2) in FIG. 6 is the same as the RECI process (S2) in FIG. 5, a duplicated explanation will be omitted. With the above, the operation of the causal relationship estimation system 2 is completed.

[0068] However, if the causal direction is estimated in advance by executing the ANM process (S1) and the RECI process (S2), S1 and S2 may be omitted.

[0069] Next, the algorithm determination process (S30) will be described with reference to FIGS. 7 and 8. When the algorithm determination process (S30) is executed, the calculation device 2b determines one determination algorithm CAa from among a plurality of different causal inference algorithms CA (S31). The determination algorithm CAa is arbitrarily selected from among a plurality of different causal inference algorithms CA. In this embodiment, a case where ANM is selected will be described as an example. However, the causal inference algorithm CA used in the algorithm determination process (S30) is not limited to ANM, and LiNGAM, PNL, etc. may also be used.

[0070] Next, after S31 is executed, the calculation device 2b executes a regression analysis process (S32), a residual calculation process (S33), and a score calculation process (S34). The processes of S32 to S34 described above are the same as the regression analysis process (S10), the residual calculation process (S11), and the score calculation process (S12) in the ANM process (S1) shown in Fig. 4, so redundant explanations will be omitted.

[0071] Next, the calculation device 2b calculates p-values ​​P for the null hypothesis that the explanatory variables and the residual R are independent, for both directions of the causal relationship (X → Y and Y → X) (S35). X→Y and p is the p-value P for when Y is the cause of X. Y→X and are calculated.

[0072] As shown in FIG. 8, the calculation device 2b calculates p X→Y and,p Y→X and obtain p X→Y and,p Y→X It is determined whether or not and are equal (S36).

[0073] The calculation unit 2b is X→Y and,p Y→XIf it is determined that and are equal (S36: Y), it is determined that the causal direction cannot be inferred (S37). The calculation device 2b assigns information indicating that inference is not possible to the causal direction value dir. For example, 0 is assigned to the causal direction value dir. This completes the algorithm determination process (S30).

[0074] The calculation unit 2b is X→Y and,p Y→X If it is determined that and are not equal (S36:N), p X→Y But, p Y→X The calculation device 2b determines whether the value is greater than p X→Y But, p Y→X If it is determined that the result is greater than 0 (S38: Y), the p-value P used as an index for selecting the causal inference algorithm CA is p i As, p X→Y is substituted (S39).

[0075] On the other hand, the calculation unit 2b calculates p X→Y But, p Y→X If it is determined that the result is not greater than p (S38:n), the p-value P used as an index for selecting the causal inference algorithm CA is p i As, p Y→X is substituted (S40).

[0076] Next, the calculation unit 2b calculates p i The calculation device 2b determines whether p is equal to or greater than the threshold value TH (S41). i is greater than or equal to the threshold value TH (S41: Y), the causal direction inferred by the ANM is adopted as the causal direction, and the causal direction value dir inferred by the ANM is substituted as the causal direction value dir (S42). This completes the algorithm determination process (S30).

[0077] On the other hand, the calculation unit 2b calculates p iIf it is determined that the causal direction is not equal to or greater than the threshold value TH (S41: N), the causal direction inferred by ANM is adopted as the causal direction, and the causal direction value dir inferred by RECI is substituted as the causal direction value dir (S43). This completes the algorithm determination process (S30).

[0078] 5. Effects of this form Next, the effects of this embodiment will be described. A causal relationship estimation system 2 according to this embodiment includes a storage device 2a that stores analysis target data DA and a plurality of different causal inference algorithms CA, and a calculation device 2b that estimates a causal relationship contained in the analysis target data DA using one causal inference algorithm CA selected from the plurality of different causal inference algorithms CA. The calculation device 2b performs a regression analysis on the analysis target data DA using one determination algorithm CAa selected from the plurality of different causal inference algorithms CA to calculate a residual R, calculates a p-value P that represents the independence between the residual R and an explanatory variable corresponding to the residual R, and uses the p-value P to determine which causal inference algorithm CA to adopt from the plurality of different causal inference algorithms CA, and estimates a causal relationship based on the adopted causal inference algorithm CA.

[0079] According to this embodiment, a residual R is calculated using a judgment algorithm CAa, and a p-value P representing the independence between this residual R and the explanatory variable corresponding to the residual R is calculated. By using this p-value P, an appropriate causal inference algorithm CA can be mechanically selected for the data DA to be analyzed.

[0080] Furthermore, the calculation device 2b according to this embodiment estimates the causal direction based on the employed causal inference algorithm CA, and estimates the causal relationship based on the employed causal inference algorithm CA and the estimated causal direction.

[0081] In addition, the calculation device 2b according to this embodiment calculates the residuals R in both directions, calculates p-values ​​P for both directions using the residuals R in both directions, determines the p-value P to be adopted using the p-values ​​P for both directions, and uses the determined p-value P to determine which causal inference algorithm CA to adopt from among a plurality of different causal inference algorithms CA.

[0082] Furthermore, according to this embodiment, the calculation device 2b compares the p-value P with a threshold value TH stored in the storage device 2a to determine which causal inference algorithm CA to adopt from among multiple different causal inference algorithms CA. According to this embodiment, an appropriate algorithm for the analysis target data DA can be mechanically selected by the simple method of comparing the p-value P with the threshold value TH.

[0083] The different causal inference algorithms CA according to this embodiment are Addetive Noise Model (ANM) and Regression Error Based Causal Inference (RECI).

[0084] As described above, ANM and RECI have different assumptions for the analysis target data DA. Therefore, when ANM and RECI are applied to one analysis target data DA to infer a causal relationship, one may be correct and the other may be incorrect. Even in such a case, according to the present embodiment, an appropriate causal inference algorithm CA for the analysis target data DA can be mechanically selected, thereby making it possible to infer a correct causal relationship.

[0085] For example, while ANM is able to estimate correct causal relationships, examples of causal relationships estimated by RECI that are incorrect include data relating to altitude and temperature, data relating to fine root decomposition, data relating to outdoor and indoor temperatures, and data relating to net ecosystem production (NEP) and photosynthetic photon density (PPFD).

[0086] Furthermore, while RECI is able to estimate correct causal relationships, examples of causal relationships estimated by ANM that are incorrect include data relating to age and hourly wages, data relating to latitude and temperature, data relating to ozone levels and temperature, data relating to the population with access to running water and infant mortality rates, data relating to clay content and organic carbon content in soil, and data relating to precipitation and runoff in river basins.

[0087] According to this embodiment, in all of the above examples, the correct causal relationship can be inferred by mechanically selecting the causal inference algorithm CA.

[0088] Furthermore, the calculation device 2b according to this embodiment applies the causal inference algorithm CA to the analysis target data DA to calculate the score C and the mean square error MSE, and infers the causal direction based on the score C.

[0089] Furthermore, according to this embodiment, it is possible to automatically select either ANM or RECI, whichever is more appropriate for the analysis target data DA.

[0090] According to this embodiment, as an independence index representing the independence of the explanatory variable corresponding to the residual R, the p-value P, which is the null hypothesis that the explanatory variable and the residual R are independent, can be used.

[0091] Furthermore, according to this embodiment, the analysis target data DA includes machine setting conditions or machining conditions of the machine tool 4 or production system 5 that executes a predetermined process on the workpiece, and data related to the processing results of executing the predetermined process on the workpiece, and the calculation device 2b infers the causal direction between the machine setting conditions or machining conditions and the processing results on the workpiece by the machine tool 4 or production system 5. This makes it possible to apply the causal relationship estimation system 2 to the machine tool 4 or production system 5.

[0092] Moreover, the production support system 1 according to this embodiment includes the above-described causal relationship estimation system 2, a server 3 or a virtual server (not shown) that stores a knowledge database DB that indicates the relationship between information about the workpiece or processing specifications and a processing result that is the result of performing a predetermined process on the workpiece, and a communication device 2e that acquires analysis target data DA by communication from the machine tool 4 or the production system 5. According to this embodiment, the causal relationship estimation system 2 can be applied to the production support system 1 that includes the knowledge database DB.

[0093] The production system 5 according to this embodiment further includes an update device 2f that updates the knowledge database DB based on the estimated causal relationships, thereby enabling the knowledge database DB to be updated based on the causal relationships of the analysis target data DA estimated using the causal relationship estimation system.

[0094] The present invention is not limited to the above-described embodiments, and can be applied to various embodiments within the scope of the present invention. [Explanation of symbols]

[0095] 1: Production support system, 2: Causal relationship estimation system, 2a: Storage device, 2b: Calculation device, 2e: Communication device, 2f: Update device, 3: Server, 4, 5c: Machine tool, 5: Production system, 6: Network, ANM: Addettive: Noise: Model, C: Score, CA: Causal inference algorithm, CAa: Judgment algorithm, CI: Causal direction index, DA: Data to be analyzed, DB: Knowledge database, dir, dir ANM ,dir RECI : Causal direction value, MSE: Mean square error, R: Residual, RECI: Regression:Error:Based:Causal:Inference, TH: Threshold

Claims

1. a storage device for storing data to be analyzed and a plurality of different causal inference algorithms; a calculation device that estimates a causal relationship included in the analysis target data using one causal inference algorithm selected from the plurality of different causal inference algorithms, The computing device performing a regression analysis on the analysis target data using one determination algorithm selected from the plurality of different causal inference algorithms to calculate a residual; calculating an independence index representing the independence between the residual and an explanatory variable corresponding to the residual; determining which causal inference algorithm to adopt from among the plurality of different causal inference algorithms using the independence index; A causal relationship inference system that infers causal relationships based on an adopted causal inference algorithm.

2. The computing device Inferring a causal direction based on the employed causal inference algorithm; The causal relationship estimation system according to claim 1 , wherein a causal relationship is estimated based on the employed causal inference algorithm and the estimated causal direction.

3. The computing device Calculating the residuals in both directions; calculating the independence index for each of the two directions using the residuals for each of the two directions; using the independence measures for both directions to determine the independence measure to be adopted; The causal relationship inference system according to claim 2 , wherein the determined independence index is used to determine which causal inference algorithm to adopt from among the plurality of different causal inference algorithms.

4. 2. The causal relationship estimation system according to claim 1, wherein the calculation device determines which causal inference algorithm to adopt from the plurality of different causal inference algorithms by comparing the independence index with a threshold value stored in the storage device.

5. The causal relationship estimation system according to claim 1 , wherein the plurality of different causal inference algorithms are Additive Noise Model (ANM) and Regression Error Based Causal Inference (RECI).

6. The causal relationship estimation system according to claim 5 , wherein the determination algorithm is the ANM.

7. The computing device applying the plurality of different causal inference algorithms to the analysis target data to calculate a causal direction index for inferring a causal direction; The causal relationship inference system according to claim 1 , wherein a causal direction is inferred based on the causal direction indicator.

8. The causal relationship inference system according to claim 1 , wherein the independence index is a p-value based on a null hypothesis that the explanatory variable and the residual are independent.

9. the analysis target data includes machine setting conditions or machining conditions of a machine tool or production system that performs a predetermined process on a workpiece, and data related to a processing result of performing the predetermined process on the workpiece; The causal relationship estimation system according to any one of claims 1 to 8, wherein the calculation device infers a causal direction between the machine setting conditions or the machining conditions and the processing results on the workpiece by the machine tool or the production system.

10. The causal relationship estimation system according to claim 9 ; a server or virtual server that stores a knowledge database that indicates the relationship between information or processing specifications related to a workpiece and a processing result that is the result of performing a predetermined processing on the workpiece; a communication device that acquires data to be analyzed from the machine tool or the production system via communication.

11. 11. The production support system according to claim 10, further comprising an update device that updates the knowledge database based on the estimated causal relationships.

Images