Control device and control method

Abstract

A control device (1A) comprises: a state acquisition unit (10) that acquires state information pertaining to a state of a machine tool which is an object to be controlled; a context information creation unit (12) that creates, on the basis of the state information, context information indicating characteristics of the operating status of the machine tool and the status of operation with respect to the machine tool; a notification information creation unit (11) that creates, on the basis of the state information, notification information which a user of the machine tool is notified of when using the machine tool; a notification determination unit (13) that determines, on the basis of the notification information and the context information, the degree of urgency of notification of the notification information; a display unit (14) that determines, on the basis of the determined degree of urgency, a display method of the notification information and that displays, by the determined display method, the notification information; and an evaluation acquisition unit (15) that acquires an evaluation from the user for the displayed notification information, wherein the notification determination unit (13) uses the context information and the evaluation to change the degree of urgency.

Description

Control device and control method

[0001] The present disclosure relates to a control device and a control method for controlling a machine tool.

[0002] It is desirable to provide various types of information to users of machine tools. Conventionally, users of machine tools have been provided with, for example, information on the operating status of the machine tool, information on any abnormalities that have occurred, information on when to replace consumables, and information on when to perform regular maintenance and inspections. Such information is often notified to users of machine tools using a display device known as an HMI (Human Machine Interface) that is provided in the control device of the machine tool. In such cases, the control device of the machine tool is required to notify the users of the machine tool in a manner that is desirable for the users of the machine tool.

[0003] For example, the control device described in Patent Document 1 sets a standby time before reporting an abnormality, and when an abnormality occurs, notifies the user of the machine tool of the abnormality after the standby time has elapsed. In this way, the control device described in Patent Document 1 notifies the user of the abnormality without interfering with the user's operation.

[0004] International Publication No. 2023 / 100364

[0005] However, in the technology of Patent Document 1, since the user is notified of an abnormality based on fixed rules, the user cannot grasp changes in status information such as how the machine tool is operating, how it is being used, etc. Therefore, although it is necessary to respond to abnormality notifications in order of urgency according to changes in status information, the user cannot prioritize the abnormality notifications and respond accordingly.

[0006] The present disclosure has been made in consideration of the above, and aims to obtain a control device that can effectively provide high-urgency notifications to the user even when the status information of the machine tool changes.

[0007] To solve the above-mentioned problems and achieve the object, a control device disclosed herein includes a status acquisition unit that acquires status information, which is information about the status of a machine tool to be controlled, and a context information creation unit that creates, based on the status information, context information, which is information that represents the operating status of the machine tool and characteristics of the operating status of the machine tool. The control device disclosed herein also includes a notification information creation unit that creates, based on the status information, notification information to be notified to a user of the machine tool when using the machine tool, and a notification determination unit that determines a level of urgency for the notification information based on the notification information and the context information. The control device disclosed herein also includes a display unit that determines a display method for the notification information based on the determined level of urgency and displays the notification information using the determined display method, and an evaluation acquisition unit that acquires a user's evaluation of the displayed notification information. The notification determination unit changes the level of urgency using the context information and the evaluation.

[0008] The control device according to the present disclosure has the effect of being able to effectively notify the user of a highly urgent event even when the status information of the machine tool changes.

[0009] FIG. 1 is a diagram showing the configuration of a control device according to a first embodiment; FIG. 2 is a diagram for explaining an example of an urgency parameter set used by the control device according to the first embodiment; FIG. 3 is a diagram showing an example of an ID system for a screen state used by the control device according to the first embodiment; FIG. 4 is a diagram showing an example of a setup screen displayed by the control device according to the first embodiment; FIG. 5 is a diagram for explaining first to third examples of a display method for notification information displayed by the control device according to the first embodiment; FIG. 6 is a diagram for explaining fourth to sixth examples of a display method for notification information displayed by the control device according to the first embodiment; FIG. 1 is a diagram for explaining an example. FIG. 2 is a flowchart showing the processing procedure of processing executed by the control device according to the first embodiment. FIG. 3 is a diagram showing the configuration of the control device according to the second embodiment. FIG. 4 is a diagram for explaining the Elbow algorithm used by the control device according to the second embodiment. FIG. 5 is a diagram for explaining vector data of TF-IDF used by the control device according to the second embodiment. FIG. 6 is a diagram for explaining clustering by the k-means method used by the control device according to the second embodiment. FIG. 7 is a diagram for explaining an example of an urgency parameter set when the control device according to the second embodiment determines the urgency using notification information clusters. FIG. 8 is a diagram for explaining an example of an urgency parameter set when the control device according to the second embodiment determines the urgency using long contexts and short contexts. FIG. 9 is a diagram for explaining processing by the control device according to the second embodiment to manage the urgency for each user using user identification information.

[0010] Hereinafter, a control device and a control method according to an embodiment of the present disclosure will be described in detail with reference to the accompanying drawings.

[0011] 1 is a diagram showing the configuration of a control device according to a first embodiment. The control device 1A is a computer that controls a machine tool (not shown). The control device 1A includes a status acquisition unit 10, a notification information creation unit 11, a context information creation unit 12, a notification determination unit 13, a display unit 14, an evaluation acquisition unit 15, and an information storage unit 16.

[0012] Note that in the first embodiment, explanations of functions provided in a general control device (machine tool control device) will be omitted. For example, explanations of functions such as a function for analyzing a machining program to create movement trajectories for axes of a machine tool, a function for creating movement speed commands for rotation axes including the spindle, and a function for outputting operation commands to motors of the spindle and servo axes will be omitted, and only the configuration and processing unique to the machine tool control device 1A of the present disclosure will be explained.

[0013] First, the role of each component included in the control device 1A and the process until the effects of embodiment 1 are achieved will be outlined. Here, detailed explanations will be omitted and an overall picture of embodiment 1 will be explained. Detailed explanations of each component included in the control device 1A and detailed explanations of important points in embodiment 1 will be given later.

[0014] The status acquisition unit 10 acquires status information, which is information on the status of the machine tool to be controlled. The status information is information that indicates how the machine tool is operating and how it is being used.

[0015] The status information includes, for example, at least one of information on the control status of the machine tool by the control device 1A, information on the operating status of the machine tool, information on the screen status of the screen displayed by the display unit 14 of the control device 1A, and information on the operating status of the control device 1A by the user (operator, etc.).

[0016] The control status information indicates what type of control the control device 1A is performing. The operation status information indicates the operating status of the machine tool. The screen status information indicates the status of the screen displayed on the display unit 14. The operation status information indicates the content of the operation performed by the user on the machine tool or the control device 1A.

[0017] Sensors, various measuring instruments, various devices, etc. (not shown) attached to the machine tool to be controlled and the control device 1A provide status information to the status acquisition unit 10. Depending on the type of status information, the status acquisition unit 10 may acquire data held by the control device 1A as status information.

[0018] The status acquisition unit 10 acquires status information at a specific period and stores the acquired status information in the information storage unit 16 so that the status information can be used by the notification information creation unit 11 and the context information creation unit 12.

[0019] The notification information creation unit 11 creates notification information based on the status information acquired by the status acquisition unit 10. The notification information is information to be notified to the user of the machine tool when the machine tool is being used. By notifying the user of the notification information, appropriate use of the machine tool is encouraged.

[0020] In order for the machine tool to be used appropriately, it is desirable that appropriate information be provided to the user from the machine tool and control device 1 A. Specifically, the notification information includes information on the current state of the machine tool (current state information), information required for automatic operation of the machine tool (automatic operation information), information on an abnormality that has occurred in the machine tool (abnormality information), and warning information associated with an operation that is not permitted for the machine tool (warning information).

[0021] The notification information creating unit 11 may create appropriate notification information corresponding to the status information using specific status information as a trigger, or may create appropriate notification information corresponding to the status information based on changes in the status information or time-series analysis. The notification information creating unit 11 outputs the created notification information to the notification determining unit 13.

[0022] The context information creating unit 12 creates context information, which is information on a context, based on the state information acquired by the state acquiring unit 10. The context information in the first embodiment is information indicating the characteristics of the operating status of the machine tool and the characteristics of the operating status of the machine tool.

[0023] In the first embodiment, the characteristics of the operating status of the machine tool may be referred to as an operating context, and the characteristics of the operating status of the machine tool may be referred to as an operating context. Furthermore, when the term "context" is used simply in the first embodiment, it refers to information that includes both of these without making a distinction between them.

[0024] The context information creation unit 12 performs various data processing on the state information. For example, the context information creation unit 12 performs statistical processing on the state information to grasp the characteristics of the state in which the machine tool is placed. The context information creation unit 12 identifies the context information corresponding to the state information by determining to which of a plurality of pieces of context information registered in advance as candidates the grasped characteristics corresponds. The context information creation unit 12 outputs the context information determined in this manner to the notification determination unit 13. Candidates for context information are stored in advance in a storage device such as the information storage unit 16.

[0025] The notification determination unit 13 determines the urgency of notification for a combination of the notification information and the context information, based on the notification information created by the notification information creation unit 11 and the context information created by the context information creation unit 12. The urgency is an index used by the display unit 14 to determine how to notify the notification information. The notification determination unit 13 outputs the determined urgency to the display unit 14. The display unit 14 determines how to display the notification information so that the higher the urgency, the more it attracts the user's attention, and so that the lower the urgency, the less it disturbs the user.

[0026] The urgency levels in the first embodiment include a level of urgency used to determine a display method for notification information and a level of urgency used to calculate this level of urgency. The levels of urgency used to calculate this level of urgency include a fixed level of urgency and a variable level of urgency. Hereinafter, the fixed level of urgency used to calculate the level of urgency may be simply referred to as a fixed value, and the variable level of urgency used to calculate the level of urgency may be simply referred to as a variable value.

[0027] 2 is a diagram illustrating an example of an urgency parameter set used by the control device according to embodiment 1. The urgency parameter set 2A includes fixed values ​​and variable values ​​that are parameters for determining the urgency.

[0028] The control device 1A calculates the urgency used to determine the display method of notification information, using fixed values ​​and variable values ​​included in the urgency parameter set 2A. The urgency parameter set 2A, which is a set of urgency parameters (fixed values ​​and variable values), is used when determining the urgency for a combination of notification information and context information. In the control device 1A, the urgency parameter set 2A is stored in advance in a storage device such as the information storage unit 16. The urgency parameter set 2A is the entire set of fixed values ​​and variable values ​​for determining the urgency managed for each notification information, each notification information cluster described below, and each context information.

[0029] As shown in Fig. 2, the urgency used to calculate the urgency has a fixed value (urgency fixed value) and a variable value (urgency variable value) for each piece of notification information. In the urgency parameter set 2A, for example, notification information having an ID (IDentification) that is an identifier of the notification information of "001" is associated with a fixed value corresponding to the notification information of "001". Also, notification information having an ID of "002" is associated with a fixed value corresponding to the notification information of "002".

[0030] Furthermore, various types of context information such as contexts Ax, Bx, and Cx are associated with notification information with an ID of "001." Similarly, various types of context information such as contexts Ax, Bx, and Cx are associated with notification information with an ID of "002." One variable value is associated with each combination of notification information and context information. For example, one variable value is associated with the combination of notification information with an ID of "001" and context Ax.

[0031] The control device 1A determines the final urgency level for each notification information based on these two values ​​(the fixed value and the variable value). The final urgency level is the urgency level at this point in time, and if the variable value is subsequently changed, the final urgency level also changes.

[0032] The fixed value is a value that is set in advance and is unique to each piece of notification information, and basically does not change. On the other hand, the variable value has a value for each combination of notification information and context information, and is updated successively according to the evaluation information described later. The fixed value and variable value will be described in detail later.

[0033] The notification determination unit 13 applies a variable value according to the context information. By switching the variable value according to the context information, the notification determination unit 13 can select the urgency level according to the usage status (status information) of the machine tool.

[0034] Furthermore, the notification determination unit 13 updates the variable value in accordance with the evaluation information output from the evaluation acquisition unit 15. This enables the control device 1A to change the urgency to the urgency desired by the user for each usage situation of the machine tool.

[0035] The notification determination unit 13 determines the urgency level using a fixed value in the initial state by setting the variable value before the user evaluates the notification information, i.e., the initial value of the variable value, to 0. This allows the notification information to be displayed for the first time. Thereafter, the user evaluates the notification information and the variable value changes, thereby updating the variable value. This allows the control device 1A to realize an operation of changing the notification information to an appropriate display according to the status information.

[0036] The display unit 14 determines a display method for the notification information according to the level of urgency, and displays the notification information using the determined display method. The display unit 14 can employ various display methods as long as the display method achieves the purpose of notifying the user of the machine tool of the notification information and can change the degree to which it attracts the user's attention according to the level of urgency, and any display method may be applied. As will be described in detail later, the display unit 14 may change the size, color, or brightness of the display of the notification information based on the level of urgency, or may change the display format using a pop-up or a small icon, etc.

[0037] The evaluation acquisition unit 15 acquires an evaluation by the user of the notification information displayed on the display unit 14. The evaluation acquisition unit 15 acquires the user's evaluation of the notification information displayed on the display unit 14 and outputs the evaluation information to the notification determination unit 13. The user's evaluation method may be any method as long as it is a system for numerically evaluating whether or not the user perceived the display method or display timing of the notification information displayed on the display unit 14 favorably.

[0038] As will be described in detail later, the display unit 14 displays, for example, a button representing a positive evaluation and a button representing a negative evaluation alongside the notification information. In this case, the evaluation acquisition unit 15 acquires the user's evaluation based on whether the button representing a positive evaluation or the button representing a negative evaluation is pressed. Alternatively, the display unit 14 may display a button for checking the notification information later alongside the notification information, and the evaluation acquisition unit 15 may accept the user's evaluation as a negative evaluation when this button is pressed.

[0039] The notification determination unit 13 updates the variable value of the combination of notification information and context corresponding to the evaluation information, based on the evaluation information output from the evaluation acquisition unit 15. This enables the control device 1A to display appropriate notification information according to the usage status of the machine tool.

[0040] Next, each of the components of the control device 1A will be described in detail.

[0041] [Status Acquisition Unit 10] As described above, the status acquisition unit 10 acquires status information of the machine tool. The status information will now be described in detail. For example, when the status information indicates the control status of the machine tool by the control device 1A, the status information includes information on which functions of the machine tool are enabled, and, if multiple modes are available for each function, information on which mode is active. Furthermore, when the status information indicates the control status of the machine tool, the status information includes information on which command is being executed in the currently running machining program. In this case, the status information includes the contents of the G code that commands the function of the control device 1A to be executed, the contents of the M code that commands various auxiliary functions, the value of the F command that commands the axis movement speed, the value of the S command that commands the spindle rotation speed, the value of the T command that selects the tool to be used for machining, and so on. Furthermore, the status information in this case includes information on the values ​​of various parameters provided in the control device 1A.

[0042] The state of the machine tool changes depending on the parameter setting value. For example, when the parameter setting value is changed, the state changes to various states, such as a state in which a specific function is always enabled, a state in which a specific function is disabled, or a state in which one of multiple modes or types is selected.

[0043] For example, when the enable parameter for a specific function F1 is on, it is natural that function F1 is enabled, and control may be performed to switch the operation of function F1 to a specific mode, mode M1 or mode M2, by referring to the value of another parameter. Furthermore, control may be performed in combination with commands from the machining program to enable or disable functions, or to switch modes. These controls change the state of the machine tool, so the state acquisition unit 10 acquires the parameter setting values ​​as state information. The state acquisition unit 10 may also acquire data used by the software of the control device 1A when executing control as state information.

[0044] Furthermore, when the status information indicates the operating status of the machine tool, the status information broadly includes information such as automatic operation, automatic pause, manual operation, stopped, and abnormal state (alarm activated).

[0045] During automatic operation, the machine tool operates automatically according to a machining program that describes the machining or operation details, and when this state is temporarily stopped, it is called automatic pause.

[0046] During manual operation, the machine tool user can manually operate the machine tool. During manual operation, the user can move the axis using a manual pulse generator, or move or rotate the axis using various buttons on the machine control panel, such as jog operation, a return-to-home button, or a fast-forward button. Such manual operation is often used during setup, such as loading and unloading a workpiece or preparing for machining, or measuring a tool or workpiece. For this reason, manual operation can be said to be an element that suggests the possibility of a state in which setup, tool or workpiece measurement, etc., is in progress.

[0047] When the machine is stopped, no operation corresponding to automatic operation or manual operation is being performed, and the machine tool is simply in a state where it is powered on. However, since there is a possibility that an operation is being performed on the display unit 14 even when the machine is stopped, the control device 1A can accurately grasp the usage status of the machine tool by also acquiring the screen status or operation status.

[0048] The abnormal state is a state in which some kind of abnormality occurs in the machine tool. In the abnormal state, for example, an alarm or warning is being output from the control device 1A.

[0049] The operating states may also include information on how long each operating state has been in. The state acquisition unit 10 can also acquire the power-on time of the machine tool as state information by calculating the sum of each operating state.

[0050] The status acquisition unit 10 may acquire these operating states from data held by the software of the control device 1A, or may acquire them from a PLC (Programmable Logic Controller) device that serves as an interface between the machine tool and the control device 1A.

[0051] Furthermore, when the status information indicates the screen status displayed by the display unit 14, the status information is information about the status of the screen currently being displayed. The control device 1A assigns identification information, such as a uniquely identifiable ID or tag, to each piece of screen data prepared in advance, so that the status acquisition unit 10 can acquire the identification information of the currently displayed screen as the screen status. Furthermore, when a screen is managed by multiple pages, the screen status may be a combination of information on the page number and the identification information of the screen.

[0052] 3 is a diagram showing an example of an ID system of screen states used by the control device according to embodiment 1. The screen state is expressed by the screen state of the main screen, the screen state of the sub-screen, and the screen state of the screen components.

[0053] The screen state of the sub-screen corresponds to the area displayed on the main screen, and the screen state of the screen parts corresponds to parts such as buttons displayed on the main screen.

[0054] The main screens include an operation screen that displays coordinate values ​​and the like while the machine tool is operating, a setup screen that is displayed while the machine tool is being set up, an edit screen for editing machining programs, a maintenance screen for changing the settings of parameters, etc. Figure 3 shows a case where the identifier of the operation screen is ID: 01 and the identifier of the setup screen is ID: 02.

[0055] The sub-screen includes a first information display area, a main screen menu button display area, etc. Fig. 3 shows a case where the identifier of the first information display area is S (Sub) ID: 001, and the identifier of the main screen menu button display area is SID: 004.

[0056] The screen components include a vertical scroll bar, a left scroll button, etc. In Fig. 3, the identifier of the vertical scroll bar is P (Part) ID: 001, and the identifier of the left scroll button is PID: 003.

[0057] Fig. 4 is a diagram showing an example of an operation screen displayed by the control device according to the first embodiment, and Fig. 5 is a diagram showing an example of a setup screen displayed by the control device according to the first embodiment. Fig. 4 shows the screen state of an operation screen 2a, and Fig. 5 shows the screen state of a setup screen 2b.

[0058] The screen state of the operation screen 2a, which is the main screen, is expressed by the screen state of the operation screen 2a, the screen state of the sub-screen corresponding to the operation screen 2a, and the screen state of the screen components. Also, the screen state of the setup screen 2b, which is the main screen, is expressed by the screen state of the setup screen 2b, the screen state of the sub-screen corresponding to the setup screen 2b, and the screen state of the screen components.

[0059] The operation screen 2a is displayed by combining sub-screens (display areas) corresponding to the operation screen 2a. Fig. 4 shows a case where the sub-screens corresponding to the operation screen 2a are a first information display area 31a, a main screen transition button display area 32a, a coordinate value display area 33a, a machining program display area 34a, a second information display area 35a, a control mode display area 36a, a third information display area 37a, and a main screen menu button display area 38a.

[0060] Also displayed on the operation screen 2a are, for example, a left button 41a, a right button 42a, a menu button 43a within the main screen, and a button 44a for switching the main screen.

[0061] The main screen transition button display area 32a is an area where buttons 44a for switching the main screen are displayed. The coordinate value display area 33a is an area where coordinate values ​​defined in the machining program, coordinate values ​​indicating the actual position of the machine tool (machine position), etc. are displayed. The machining program display area 34a is an area where the machining program is displayed. The control mode display area 36a is an area where the control mode of the control device 1A is displayed. The main screen menu button display area 38a is an area where menu buttons 43a within the main screen are displayed. The first information display area 31a, the second information display area 35a, and the third information display area 37a are areas where information about the machine tool is displayed.

[0062] The setup screen 2b is displayed by combining sub-screens (display areas) corresponding to the setup screen 2b. Fig. 5 shows a case where the sub-screens corresponding to the setup screen 2b are a first information display area 31b, a main screen transition button display area 32b, a coordinate value display area 33b, a second information display area 35b, a third information display area 37b, and a main screen menu button display area 38b. The second information display area 35b includes a parameter display area 39b and an input data display area 40b.

[0063] The setup screen 2b also displays, for example, a left button 41b, a right button 42b, a menu button 43b within the main screen, a button 44b for switching the main screen, a vertical scroll bar 45b, and the like.

[0064] The main screen transition button display area 32b is an area where buttons 44b for switching the main screen are displayed. The coordinate value display area 33b is an area where coordinate values ​​defined in the machining program, coordinate values ​​indicating the actual machine position of the machine tool, etc. are displayed. The main screen menu button display area 38b is an area where menu buttons 43b within the main screen are displayed. The parameter display area 39b is an area where various parameters are displayed. The input data display area 40b is an area where data entered by the user is displayed. The first information display area 31b and the third information display area 37b are areas where information about the machine tool is displayed.

[0065] In the control device 1A, when sub-screens can be displayed by combining multiple sub-screen display areas in the form of components on a single main screen, unique identifiers (SIDs) are also assigned to these sub-screens. This allows the status acquisition unit 10 to combine multiple SIDs with the currently displayed screen state. For example, in FIG. 4, sub-screens such as a first information display area 31a, a coordinate value display area 33a, a machining program display area 34a, and a second information display area 35a are combined with an operation screen 2a, which is the main screen. In this case, the control device 1A can represent the screen state with the ID of the operation screen 2a and the SID of the sub-screen.

[0066] Furthermore, the screen state is further subdivided, and a unique identifier (PID) is assigned to each screen component that constitutes the screen. This allows the state acquisition unit 10 to combine these PIDs with the screen state currently being displayed. For example, in FIG. 5, a sub-screen of a main screen menu button display area 38b is combined with a main screen setup screen 2b, and further screen components such as a left scroll button 41b, a right scroll button 42b, and a vertical scroll bar 45b are combined. In this case, the control device 1A can represent the screen state using the ID of the setup screen 2b, the SID of the sub-screen, and the PID of the screen component.

[0067] To obtain a more detailed ID representation of the screen state, the control device 1A may combine the screen state with coordinate values ​​within the screen. In this case, the control device 1A may use a screen state that includes, for example, the position at which a sub-screen or screen component is displayed. The control device 1A may also represent the screen state using a hierarchical ID structure using a nested structure between sub-screens or between a sub-screen and a screen component. In this case, the control device 1A provides a hierarchical ID structure, for example, as shown in FIG. 5 , by providing a parameter display area 39b within the second information display area 35b and further providing a vertical scroll bar 45b within the second information display area 35b.

[0068] In recent years, there are cases where each machine tool manufacturer uses its own custom screen. In such cases, for example, a unique ID is assigned to each machine tool manufacturer. This allows the machine tool manufacturer to assign IDs to each screen it has developed as child IDs, with the machine tool manufacturer's unique ID as the parent, making it possible for the control device 1A to support screens from various machine tool manufacturers.

[0069] If the display unit 14 has a touch panel function, the display unit 14 represents a touch operation on the display unit 14 as a coordinate value where the input is sensed. The state acquisition unit 10 may apply this coordinate value to data on the operation status. Alternatively, the state acquisition unit 10 may convert the coordinate value into operation content indicating which operation was performed by combining the coordinate value with the screen status, and apply an identifier uniquely assigned to this operation content to the data on the operation status.

[0070] Examples of the operation state include a transition operation for turning pages on a screen having multiple pages, and an operation for selecting a specific parameter on a parameter screen that displays parameters used by a machine tool.

[0071] Furthermore, in the case of an operation on a physical switch or the like, the status acquisition unit 10 may apply interface data that converts the operation content of this operation into data that can be handled by the control device 1A as the data on the operation status.

[0072] For example, when a user rotates a manual pulse generator, the rotation operation is input to the control device 1A as a pulse signal, and the status acquisition unit 10 can apply the input amount, cumulative value, maximum value, etc. of this pulse signal over a certain period of time to the operation status data.

[0073] Furthermore, when a button is pressed, the state acquisition unit 10 can apply the off or on state of the contact to the operation state data as binary data of "0" or "1." Furthermore, in the case of a switch having multiple states such as a toggle switch, the state acquisition unit 10 can apply the off or on state of the corresponding contact to the operation state data by treating it as binary data of "0" or "1."

[0074] The status acquisition unit 10 acquires the various types of status information described above at regular time intervals. The shorter the time interval (acquisition interval) at which the status acquisition unit 10 acquires the status information, the more detailed the changes in the status information can be captured by the control device 1A. This reduces the possibility that the control device 1A will miss a sudden change in the status information.

[0075] Furthermore, by extending the interval at which the status information is acquired by the status acquisition unit 10, the control device 1A can reduce the amount of data acquired and also simplify the acquisition process, thereby reducing the acquisition processing time. The user can arbitrarily set how the status information acquisition interval is set for the status acquisition unit 10.

[0076] The control device 1A may change the interval at which status information is acquired for each type of status. For example, since the control status may be changed at short intervals of milliseconds by software processing, the control device 1A may set the interval at which status information for the control status is acquired to be shorter than the interval at which other status information is acquired. On the other hand, since the operation status is a manual operation by the user, the control device 1A may set the interval at which status information for the operation status is acquired to be around one second.

[0077] The control device 1A not only acquires the current state obtained moment by moment as state information in this way, but can also define the state information as a state obtained by statistically processing the state over a certain period of time. For example, the control device 1A may use the most frequent state over a certain period of time or the state that has been maintained for the longest time as state information. In this case, the control device 1A can prevent momentary changes in the state from becoming disturbances and can acquire stable state information. In the first embodiment, there are no particular restrictions on how the state information is handled over time, and it can be changed as appropriate.

[0078] The status acquisition unit 10 stores the acquired status information in the information storage unit 16. This information storage unit 16 may be any storage device accessible from the control device 1A. The status acquisition unit 10 may be provided in the control device 1A, or may be an external storage device connected to the control device 1A so as to be able to communicate with it by any of various connection methods, whether wired or wireless.

[0079] [Notification information creation unit 11] As described above, notification information creation unit 11 creates notification information to be referenced when using the machine tool, based on the status information acquired by status acquisition unit 10. In order for the machine tool to be used appropriately, it is desirable that appropriate notification information for encouraging appropriate use of the machine tool be provided from the machine tool and control device 1A.

[0080] Current status information (information regarding the current status of the machine tool), which is an example of notification information, includes, for example, information regarding the remaining amount of consumables such as coolant or lubricant, information regarding the lifespan of tools, information regarding tool wear, and information regarding the status of drive parts such as motors or ball screws.

[0081] If the conditions of these machine tools are not maintained appropriately, not only will they not be able to fully perform as machine tools, but they may also cause breakdowns. For this reason, the notification information notified by the control device 1A may include, in addition to current status information of the machine tool, information that plays a role in preventive maintenance to prevent breakdowns (preventive information). For example, the control device 1A compares a value indicating the status of the machine tool with a reference status value, and includes in the notification information preventive information that prompts replacement or replenishment of consumables, etc., based on the comparison result.

[0082] Other examples of current status information include inspection information for devices or equipment that require inspection at regular intervals. The inspection information includes, for example, information on whether or not periodic inspections are being carried out daily, weekly, monthly, or annually, advance notice of the inspection date, and warning that the inspection date will be exceeded.

[0083] Examples of automatic operation information, which is an example of notification information, include information regarding the start and end of machining, information regarding the machining program used to operate the machine tool, information regarding changes in operation mode or control mode during operation, and information regarding cooperation with peripheral devices such as robots or pallet pool systems.

[0084] When machining by the machine tool is completed, the user must remove the machined workpiece and replace it with the next workpiece to be machined, etc. In this case, information regarding the start and end of machining is notified to the user as notification information, allowing the user to easily perform the appropriate work.

[0085] Furthermore, when an automated system is constructed in cooperation with peripheral devices such as a robot or a pallet pool system, information that will act as a trigger for starting the next operation of these peripheral devices is required. In this case, by notifying the user of information relating to cooperation with the peripheral devices as notification information, the user can easily carry out the appropriate work.

[0086] Examples of notification information, such as abnormality information (information about an abnormality that has occurred) or warning information (information about a warning associated with an unauthorized operation), include, for example, the following information: - Information on improper operation: Information indicating that the combination of operations performed by the user or the combination of functions used is improper, information indicating that a protective function for preventing damage to the machine tool or the devices that make up the machine tool has been activated, information indicating that the conditions for a function to operate have not been met, etc. - Information on the cause of improper operation: Information indicating the cause of the improper operation when the above-mentioned improper operation occurs - Information on hardware abnormalities: Information on abnormalities due to failure of the motor or detector, etc., information on connection abnormalities, information on abnormalities in power supply, information on abnormalities in electronic components (circuit boards, memory, CPU (Central Processing Unit), etc.) - Information on software abnormalities: Information on abnormalities that have occurred in various communications, information on abnormalities in parameters or setting values ​​commanded to machining programs, information on abnormalities in calculation processes and processing times within the system, information on abnormalities in the verification or authentication of user information, authority, passwords, etc. - Expiration date information: Information on a pre-set expiration date, information on the lifespan of the tool or battery, information on the timeout period when waiting for a response, etc. - Capacity information: Information on battery capacity, information on memory capacity, information indicating that the number of various registrations has reached or is approaching its upper limit, etc. - Information on the status of peripheral devices: information on the status of hydraulic pressure, information on the status of pump pressure, information on the status of lubricating oil, information on the status of coolant (remaining amount, shortage, etc.), information on abnormalities in the power supply system, information on abnormalities in the temperature sensor, etc. - Information on the status of the system: information on the process being performed by the machining system including the machine tool, information on waiting for another operation, information on when a function is enabled, information on when a function is disabled, etc.

[0087] When an abnormality occurs, the machine tool is likely to stop operation, and it is necessary to eliminate the cause and resolve the abnormality in order to prevent operation from being stopped for a long period of time. In this case, the above-mentioned abnormality information or warning information is notified to the user as notification information, so that the user can easily carry out the work to resolve the abnormality.

[0088] In addition, when an unauthorized operation is performed, the control device 1A may notify the user by adding information appropriate to the content, such as which operation is not permitted, the reason why it is not permitted, and what alternative means are available, to the notification information, thereby encouraging the user to use the device appropriately.

[0089] The notification information creation unit 11 creates the above-mentioned notification information based on the status information. The notification information creation unit 11 may create the notification information using specific status information as a trigger, based on a time-series analysis of the status information, or based on changes in the status information, but other methods are also possible. The notification information creation unit 11 may create the notification information using any method as long as the method is based on status information that can be acquired by the status acquisition unit 10.

[0090] In a method of creating notification information using specific status information as a trigger, the status information used as the trigger is registered in advance in the control device 1A. In this method, when the registered status information is acquired, the notification information creation unit 11 creates notification information corresponding to this status information. For example, the control device 1A stores information on the elapsed time since the last inspection as information on the timing of the periodic inspection, and creates notification information related to the periodic inspection when the actual power-on time since the last inspection has exceeded the stored elapsed time. Furthermore, when an abnormal state occurs, the control device 1A may create notification information corresponding to this abnormal state.

[0091] In a method for generating notification information based on a time-series analysis of status information, the target status information is registered in advance in the control device 1A. In this method, the control device 1A monitors the temporal transition of the registered status information and generates corresponding notification information when the status information exceeds a threshold or when the transition of the status information forms a specific trend pattern. For example, the control device 1A monitors the remaining amount of coolant or lubricant and generates notification information indicating a low remaining amount when the remaining amount falls below a threshold. The control device 1A may also monitor the motor current or load value and compare the trend pattern of the monitored value with failure trend data obtained in advance through experiments or the like, and generate information notifying of a sign of failure when the trend pattern of the monitored value becomes similar to the failure trend data.

[0092] In the method of creating notification information based on changes in status information, the control device 1A does not perform processing as complex as the time-series analysis described above. Instead, it detects changes in status information that have been set as targets in advance and creates notification information corresponding to the detected changes. This allows the control device 1A to create notification information by specifying specific status changes, not just simple changes in status information, by specifying status information before and after the change. For example, if the condition for creating notification information is a change in the operating state from automatic operation to stopped, the control device 1A creates notification information required at the time of machining completion. In this case, the control device 1A creates notification information related to the next task or work instructions, such as removing, measuring, or cleaning the workpiece after machining has been completed. As described above, the notification information creation unit 11 creates notification information using the status information acquired by the status acquisition unit 10.

[0093] [Context information creation unit 12] As described above, the context information creation unit 12 creates context information, which is information on context, based on the state information acquired by the state acquisition unit 10. The contexts in the first embodiment include an operation-related context indicating the characteristics of the operating status of the machine tool, and an operation-related context indicating the characteristics of the operating status of the machine tool.

[0094] The operation system context includes the operating status of the machine tool, i.e., the operating status or control state. The operation system context includes the status of operation on the display screen or the machine control panel. The operation system context also includes the status of operation on the machine tool, the status of operation on the control device 1A, and the status of operation on peripheral devices.

[0095] The context information creation unit 12 performs various data processing on the state information. For example, the context information creation unit 12 performs statistical processing on the state information to grasp the characteristics of the state in which the machine tool is placed, and determines which of a plurality of context information registered in advance as candidates the state information corresponds to.

[0096] The data processing executed by the context information creating unit 12 includes, for example, analyzing the frequency of appearance of various types of state information within a certain period of time and determining state information with a high frequency as context information.

[0097] In addition, the context information creation unit 12 may correspond one piece of state information (e.g., the most frequently occurring state information) to the context information, or may determine the context information from a plurality of pieces of state information (e.g., the top three pieces of state information in order of frequency) according to a predetermined rule or arithmetic formula.

[0098] The context information creating unit 12 may perform data processing other than that described above. For example, the context information creating unit 12 may perform another data processing of determining context information based on the magnitude of variation in the appearance frequency of state information over a certain period of time. Furthermore, the context information creating unit 12 may perform another data processing of assigning a weight to each type of state information and determining context information based on the magnitude of a value obtained by multiplying the weight by the frequency over a certain period of time.

[0099] Here, specific examples of context information will be described. For example, if the operating state of the state information includes "automatic operation," the action context may become "automatic operation" as a result of data processing by the context information creation unit 12. At this time, if there is no change in the screen state of the state information, or if nothing is recorded in the operation state, the operation context may become "abandoned state" as a result of data processing by the context information creation unit 12. From these two pieces of context information, it can be inferred that the machine tool is automatically performing the instructed machining, and that the user is not operating it or is not near the machine tool.

[0100] Furthermore, if the operating state in the state information includes "manual operation" and "stopped," and the context information creation unit 12 compares the frequency over a certain period as data processing and finds that "stopped" is dominant, the operation context may be "stopped." In this case, if many changes are recorded in the screen state in the state information and many operations are also recorded in the operation state, the operation context may be "working state" as a result of data processing by the context information creation unit 12. Here, if "many changes" are recorded in the screen state, this means that the screen state has changed more frequently than the standard frequency of changes in the screen state over a predetermined period (a reference value for the number of changes). Furthermore, if "many operations" are recorded in the operation state, this means that more operations have been performed than an index of the number of operations over a predetermined period (a reference value for the number of operations).

[0101] When the action context is "stopped" and the operation context is "working," these two pieces of context information (action context and operation context) suggest that the machine tool has stopped machining and the user is performing some kind of work.

[0102] Furthermore, the operating state of the state information may include an "abnormal state," and the control state may include information about an alarm that is occurring and information about an enabled function (enabled function). In this case, the context information creation unit 12 performs data processing, for example, by comparing the content of the alarm information with information about enabled functions to extract a function that is estimated to be the cause of the alarm. For example, if the content of the alarm information is "high-precision control in progress and commands cannot be issued," the information about enabled functions may include the "high-precision control" function and function F2, the combination of which with the "high-precision control" function is prohibited by the specifications. In other words, if "high-precision control in progress and commands cannot be issued," the combination of the "high-precision control" function and function F2 may be included. In such a case, the context information creation unit 12 compares the content of the alarm information with information about enabled functions to estimate that function F2 is the cause of the alarm.

[0103] In this way, the context information creation unit 12 processes data by combining the operating state and the control state, so that the operational context can become "an alarm is occurring in the function being used," or more specifically, "an alarm is occurring due to function F2."

[0104] In this way, the context information creating unit 12 can obtain context information that expresses the characteristics of the situation in which the machine tool is placed by performing various data processing on the state information.

[0105] The context information creating unit 12 may handle the action context and the operation context separately and output a set of the two contexts as context information. Alternatively, the context information creating unit 12 may determine a final context from the action context and the operation context and output the determined final context as a single piece of context information.

[0106] For example, if the action context is "automatic operation" and the operation context is "abandoned state," the context information can be expressed as "unmanned operation." Also, if the action context is "stopped" and the operation context is "working state," the context information can be expressed as "preparation in progress."

[0107] The context information data is expressed by defining candidate context information in advance. In this case, by assigning an identifier such as a unique number to each piece of context information, the context information data can be made data that can be handled by the control device 1A. By applying such an expression method, all possible context information can be defined, and it is possible to prevent undefined context information from appearing as a result of data processing. For example, by defining a matrix in which the rows are assigned IDs of action-related contexts and the columns are assigned IDs of operation-related contexts, and by setting the ID of the final context information as an element corresponding to the combination of rows and columns, the context information creation unit 12 can create context information.

[0108] [Notification determination unit 13] As described above, the notification determination unit 13 determines the urgency level for each context of the notification information created by the notification information creation unit 11. The urgency level has a fixed value and a variable value for each notification information as shown in Fig. 2, and the notification determination unit 13 determines the final urgency level for the notification information based on these two values. The fixed value and the variable value will be described later.

[0109] Here, a specific example of determining the urgency level according to a combination of notification information and context information will be described. Consider a case where, for example, preventive information from the notification information indicates that the remaining amount of coolant, a consumable item, is low. Coolant is sprayed at points where the tool and workpiece come into contact and in the machining tank of the machine to remove chips and heat generated during machining. If the coolant runs out during machining, the chips and heat will not be properly removed, which could result in a decrease in machining quality and increased wear or damage to the tool and workpiece.

[0110] Now, consider a case where the context information is the aforementioned "unmanned operation." During unmanned operation, machining is currently being performed and no operation is being performed. If left unattended, the coolant may run out, potentially resulting in the above-mentioned problem. Therefore, in this case, it is desirable to increase the level of urgency and display the message in a way that attracts attention (for example, in the case of Figure 8 described below, a full-screen display with an urgency level of 10 or higher) so that the current status can be easily noticed, even from a distance.

[0111] On the other hand, consider the case where the context information is "in setup" as mentioned above. During setup, no machining is currently being performed, so the possibility that the decrease in coolant liquid will become an immediate problem is low. Also, since the user is currently working on the machine, it is desirable to display the message with a low level of urgency so as not to interfere with the work (for example, in the case of Figure 8, the urgency is set to 0 or less and displayed as an icon).

[0112] As another example of notification information, consider the case where the notification includes inspection information indicating that a weekly inspection has not been performed. It goes without saying that regular inspections are desirable to maintain the health of machinery. However, when machining is performed continuously for a long period of time in an automated factory, it is more efficient to perform a thorough inspection before starting machining, rather than prioritizing inspections to the point of stopping the machining.

[0113] On the other hand, if the inspection is performed before machining starts, it is better to perform the inspection when the inspection cycle arrives. Therefore, it is desirable to lower the urgency if the context information is "unmanned operation" and to raise the urgency if it is "in setup." As mentioned above, even for the same notification information, the urgency is determined differently depending on the context information.

[0114] The fixed value of the urgency is the value of the urgency that the notification information is generally expected to have, and is a design value at the design stage of the notification information. For example, notification information regarding an abnormality or warning is expected to have a high urgency, and conversely, information regarding inspection is expected to have a relatively low urgency. However, as mentioned above, there are a wide variety of usage situations for machine tools, so there is a possibility that the expected urgency may differ from the urgency desired by the user.

[0115] Therefore, the notification determination unit 13 of the first embodiment sets a variable value for the urgency level and adjusts the final urgency level value by changing the variable value according to the evaluation as described below. The notification determination unit 13 initially determines the urgency level using a fixed value and displays the notification information for the first time. Thereafter, the notification determination unit 13 changes the variable value according to the evaluation, thereby changing the display method to an appropriate one according to the evaluation.

[0116] The notification determination unit 13 may use any method for determining the final urgency level as long as it uses a fixed value and a variable value. For example, the notification determination unit 13 may simply determine the sum of the fixed value and the variable value as the final urgency level, or may compare the two values ​​and determine the maximum value as the final urgency level. The notification determination unit 13 may also determine the final urgency level by substituting the fixed value and the variable value into a specific function. Updating of the variable value using the evaluation information output from the evaluation acquisition unit 15 will be described later.

[0117] [Display Unit 14] As described above, the display unit 14 determines the display method of the notification information according to the urgency level and displays the notification information in the determined display method. An example of a specific display method of the notification information by the display unit 14 will be described.

[0118] 6 is a diagram illustrating first to third examples of a display method for notification information displayed by the control device according to the first embodiment. The first to third examples of the display method by the display unit 14 are methods for changing the display size of the notification information depending on the level of urgency. That is, in the first to third examples of the display method, the display unit 14 displays the notification information larger, for example, as the level of urgency increases.

[0119] For example, when the urgency level is higher than a first reference value, the display unit 14 displays the notification information in a first example of a display method. In the first example of the display method, the display unit 14 displays the notification information 51 on the screen sC1 in a size that covers the entire screen sC1 (full screen) so as to pop up on the screen sC1, thereby urging the user to take an urgent action.

[0120] On the other hand, as the urgency level decreases, the display unit 14 reduces the size of the notification information 52 to fit in a corner area of ​​the screen sC2, thereby displaying the notification information smaller as the urgency level decreases.

[0121] For example, when the value is equal to or less than the first reference value and higher than the second reference value (< the first reference value), the display unit 14 displays the notification information in the second example of the display method. In the second example of the display method, the display unit 14 displays notification information 52, which is smaller than the notification information 51, at the forefront of the screen so as to pop up on the screen as shown in screen sC2.

[0122] Furthermore, for example, when the urgency level is equal to or lower than a second reference value, the display unit 14 displays notification information in a third example of the display method. In the third example of the display method, the display unit 14 displays a small exclamation mark icon 53 or the like in place of notification information in a specific area on the screen sC3. When the exclamation mark icon is pressed, the display unit 14 displays the notification information.

[0123] FIG. 7 is a diagram illustrating fourth to sixth examples of a display method for notification information displayed by the control device according to the first embodiment.

[0124] The fourth to sixth examples of the display method by the display unit 14 are methods for changing the display format (thickness, etc.) of the border of the area displaying the notification information depending on the level of urgency. In the fourth to sixth examples of the display method, the display unit 14 displays the notification information with a thicker border, for example, the higher the level of urgency.

[0125] For example, when the urgency level is higher than a third reference value, the display unit 14 displays the notification information according to the fourth example of the display method. In the fourth example of the display method, the display unit 14 displays the notification information on the screen sC4 with a thicker border than on the screens sC5 and sC6. That is, in the fourth example of the display method, the display unit 14 displays the notification information 54 so that the border of the notification information 54 is thicker than the borders of the notification information 55 and 56 on the screens sC5 and sC6. Furthermore, the display unit 14 flashes the border of the notification information 54 as in the screen sC4.

[0126] Furthermore, the display unit 14 displays the notification information in a fifth example of the display method when, for example, the urgency level is equal to or less than the third reference value but higher than the fourth reference value (< the third reference value). In the fifth example of the display method, the display unit 14 displays the notification information on the screen sC5 with a border that is thinner than that on the screen sC4 and thicker than that on the screen sC6. That is, in the fifth example of the display method, the display unit 14 displays the notification information 55 so that the border of the notification information 55 is thinner than that of the notification information 54 on the screen sC4 and thicker than that of the notification information 56 on the screen sC6.

[0127] Furthermore, for example, when the urgency level is equal to or lower than a fourth reference value, the display unit 14 displays the notification information in a sixth example of the display method. In the sixth example of the display method, the display unit 14 displays the notification information 56 in a banner format at the top or bottom of the screen sC6. The display unit 14 displays the notification information 56 so that the border of the notification information 56 is thinner than the borders of the notification information 54 and 55 on the screens sC4 and sC5.

[0128] Furthermore, in the fourth to sixth examples of the display method, the display unit 14 may use different colors for the notification information depending on the level of urgency. In this case, for example, the higher the level of urgency, the brighter the color the display unit 14 displays the notification information. Since bright colors generally attract attention and dark colors do not, the display unit 14 displays the frame portion displaying the notification information or the entire notification information using a bright primary color such as red or yellow when the level of urgency is higher than a fifth reference value. In this case, the display unit 14 displays the frame portion displaying the notification information or the entire notification information by decreasing the saturation or brightness of the color as the level of urgency decreases. For example, when the level of urgency is equal to or lower than a specific value (fifth reference value), the display unit 14 displays the notification information using a color such as gray, thereby changing the degree to which the notification information attracts the user's attention without changing elements other than the color.

[0129] The display unit 14 may determine the display method by using one of the above-described examples of the display method alone, or may determine the display method by combining a plurality of the examples.

[0130] Various methods for determining the display method according to the urgency level are possible, and any of these methods may be used in the first embodiment. For example, the display unit 14 may divide the range between the maximum and minimum possible urgency levels by one or more thresholds and determine the display method based on which range the urgency level falls within, or may use the urgency level value itself to determine the display method. The former is a discrete determination method, while the latter is a continuous determination method. For example, the former is suitable for discrete changes in multiple patterns, such as changing the display format, while the latter is suitable for continuous changes, such as changing the brightness, color, frequency, size, or border thickness of the display.

[0131] Here, a first example of a method for determining a display method according to the urgency level will be described. Fig. 8 is a diagram for explaining a first example of a relationship between the urgency level and the display method used by the control device according to the first embodiment.

[0132] The horizontal axis of the graph shown in Fig. 8 represents the urgency level, and the vertical axis represents the display method. Fig. 8 shows a first relationship between the urgency level and the display method. In the determination method using the first relationship in Fig. 8, the display method of the notification information is set discretely with respect to the urgency level. In other words, the determination method using the first relationship in Fig. 8 is a method of determining discrete notification information. In the determination method using the first relationship in Fig. 8, a display method is assigned to each range of an area divided into 5-level increments of the urgency level.

[0133] The display unit 14 displays the notification information in full screen when the urgency level is higher than 10. Furthermore, the display unit 14 displays the notification information in a pop-up format when the urgency level is higher than 5 and equal to or lower than 10. Furthermore, the display unit 14 displays the notification information in a banner format when the urgency level is higher than 0 and equal to or lower than 5. Furthermore, the display unit 14 displays the notification information in an icon format when the urgency level is equal to or lower than 0. For example, when the urgency level is 2.5, the display unit 14 determines that the notification information should be displayed in banner format (banner display).

[0134] FIG. 9 is a diagram illustrating the relationship between the level of urgency and the thickness of the display frame used by the control device according to the first embodiment. The horizontal axis of the graph shown in FIG. 9 represents the level of urgency, and the vertical axis represents the thickness of the display frame. FIG. 9 illustrates the relationship between the level of urgency and the thickness of the display frame. In the determination method using the relationship in FIG. 9, the thickness of the display frame of the notification information is set continuously with respect to the level of urgency. In other words, the determination method using the relationship in FIG. 9 is a method for determining continuous notification information. In this determination method, the thickness of the display frame is determined directly from the value of the level of urgency. For example, if the level of urgency is 2.5, the display unit 14 determines the thickness of the frame of the notification information to be a thickness corresponding to the level of urgency of 2.5 (2.5 pt in FIG. 9).

[0135] Next, a second example of a method for determining a display method according to the urgency level will be described. Fig. 10 is a diagram for explaining a second example of a relationship between the urgency level and the display method used by the control device according to the first embodiment.

[0136] The horizontal axis of the graph shown in Fig. 10 represents the urgency level, and the vertical axis represents the display method. Fig. 10 shows a second relationship between the urgency level and the display method. In the determination method using the second relationship in Fig. 10, the display method of the notification information is set discretely with respect to the urgency level, as in the determination method using the first relationship in Fig. 8. In other words, the determination method using the second relationship in Fig. 10 is also a method of determining discrete notification information. In the determination method using the second relationship in Fig. 10, the correspondence relationship between the display method and the urgency level value is expressed in the form of a function other than a linear function.

[0137] The display unit 14 determines the display method by inputting the value of the urgency into the function shown in Fig. 10. When the urgency is in the first range X1, the display unit 14 displays the notification information as an icon. When the urgency is in the second range X2, the display unit 14 displays the notification information as a banner. When the urgency is in the third range X3, the display unit 14 displays the notification information as a pop-up. When the urgency is in the fourth range X4, which is higher than the maximum value of the third range X3, the display unit 14 displays the notification information on the full screen.

[0138] By using the function shown in FIG. 10, the display unit 14 can set areas (first range X1, fourth range X4) in which the display method changes gradually in response to changes in urgency, and areas (second range X2, third range X3) in which the display method changes suddenly.

[0139] An example of a physical device that realizes the function of the display unit 14 is a display device such as a liquid crystal panel that is integrated with the control device 1A. Note that the physical device that realizes the function of the display unit 14 may be any device. For example, the display unit 14 may be a device that allows wired or wireless communication between the control device 1A and the display unit 14. In this case, the display unit 14 may be located at a position away from the control device 1A.

[0140] The display unit 14 may be a display device such as a touch panel having a touch function for detecting a contact position, or may be a device capable of wired or wireless communication with the control device 1A, such as a smartphone or a tablet.

[0141] [Evaluation Acquisition Unit 15] As described above, the evaluation acquisition unit 15 acquires the user's evaluation of the notification information displayed by the display unit 14, and outputs the evaluation information to the notification determination unit 13. The notification information may be evaluated by any method as long as the method numerically evaluates whether or not the user perceived the display method or display timing of the notification information displayed by the display unit 14 favorably.

[0142] Fig. 11 is a diagram for explaining a first example of a process in which the control device according to the first embodiment acquires an evaluation. Fig. 12 is a diagram for explaining a second example of a process in which the control device according to the first embodiment acquires an evaluation. Fig. 13 is a diagram for explaining a third example of a process in which the control device according to the first embodiment acquires an evaluation.

[0143] 11 to 13 show examples in which buttons for receiving evaluations are displayed in the notification information display areas of screens sC7 to sC9, respectively. For example, as shown in FIG. 11, the display unit 14 displays a "Helpful" button 61 and a "Not Helpful" button 62 on the screen sC7 along with the notification content. When the "Helpful" button 61 is pressed by the user, the evaluation acquisition unit 15 outputs evaluation information corresponding to a positive evaluation. On the other hand, when the "Not Helpful" button 62 is pressed by the user, the evaluation acquisition unit 15 outputs evaluation information corresponding to a negative evaluation.

[0144] 12, the display unit 14 may display a "Respond Now" button 63 and a "Respond Later" button 64 on the screen sC8 along with the notification content. In this case, when the "Respond Now" button 63 is pressed by the user, the evaluation acquisition unit 15 outputs evaluation information corresponding to a positive evaluation. On the other hand, when the "Respond Later" button 64 is pressed by the user, the evaluation acquisition unit 15 outputs evaluation information corresponding to a negative evaluation.

[0145] 13, the display unit 14 may display a "Hide similar notifications" button 65 on the screen sC9 along with the notification content. In this case, when the "Hide similar notifications" button 65 is pressed by the user, the evaluation acquisition unit 15 outputs evaluation information corresponding to a negative evaluation. In this way, the evaluation acquisition unit 15 acquires an evaluation corresponding to the user's evaluation behavior and outputs the evaluation information.

[0146] Here, we will explain what specific values ​​the evaluation acquisition unit 15 uses when quantifying the evaluation. Since the evaluation value is used to determine the urgency level, the evaluation unit (step size), maximum value, and minimum value must be set according to the method for determining the urgency level. Examples of the evaluation unit may be integer values ​​in increments of 1, or real numbers. When the evaluation unit is a real number, the number of digits of the real number is set.

[0147] In the first embodiment, there are no limitations on how the evaluation value is set, as long as it can determine the level of urgency. For example, ±1 is set as the minimum unit of evaluation for the evaluation acquisition unit 15. In this case, each time the button corresponding to the above-described positive evaluation or the button corresponding to the negative evaluation is pressed, the evaluation acquisition unit 15 outputs evaluation information of +1 or -1. That is, when the button corresponding to a positive evaluation is pressed, the evaluation acquisition unit 15 outputs evaluation information of +1 to the notification determination unit 13, and when the button corresponding to a negative evaluation is pressed, the evaluation acquisition unit 15 outputs evaluation information of -1 to the notification determination unit 13. In this case, the notification determination unit 13 can use the value of the evaluation information to determine the level of urgency by adding it directly to the variable value.

[0148] As another example of a method for determining the urgency level, the evaluation acquisition unit 15 may change the evaluation value in the case of a specific evaluation. For example, if the aforementioned "Hide similar notifications" button 65 is pressed, it is inferred that the user's evaluation is more negative than usual. In such a case, the evaluation acquisition unit 15 may change the urgency level significantly, for example, by changing the evaluation to -5 instead of the minimum unit of -1. Furthermore, the evaluation acquisition unit 15 may set a fixed value, such as -5, for each specific evaluation, or may set the evaluation at a specific multiplier relative to the normal evaluation.

[0149] In addition, the evaluation acquisition unit 15 may output evaluation information each time a user performs an evaluation action, as described above, or may count evaluation actions by content and output corresponding evaluation information when the number of evaluation actions exceeds a certain number.

[0150] Next, the relationship between the components of the control device 1A and the operation of each component will be described. One of the important processing flows in the first embodiment is the processing flow of the urgency determination in the notification determination unit 13.

[0151] As described above, the notification determination unit 13 determines the urgency of the target notification information using the notification information and context information. The notification determination unit 13 also uses evaluation information to update the urgency as needed. The determination and update of the urgency by the notification determination unit 13 is important for achieving the effects of the first embodiment, and will be described in detail below.

[0152] First, the data structure of the urgency level will be described with reference to Fig. 2. As described above, the notification determination unit 13 determines the urgency level based on a fixed value and a variable value. The notification determination unit 13 manages these fixed values ​​and variable values ​​as parameters that differ for each piece of notification information.

[0153] Of the fixed values ​​and the variable values, the variable values ​​are parameters that are changed or updated based on the evaluation information. In order to reflect the evaluation according to the context information, it is desirable that the variable values ​​be managed as different parameters for each piece of context information.

[0154] As mentioned above, the fixed value is the urgency assumed when the notification information is designed, so there is no need to have it for each piece of context information. However, if you want to perform more detailed notification management, the fixed value may also be managed as a different parameter for each piece of context information.

[0155] 14 is a flowchart showing the processing procedure of the processing executed by the control device according to embodiment 1. Here, the processing executed by the control device 1A will be described: the processing for creating context information and notification information, the processing for determining the urgency level, the processing for determining the display method, and the processing for updating the urgency level based on the evaluation.

[0156] The status acquisition unit 10 acquires status information, which is information on the status of the machine tool to be controlled (step S1). The context information creation unit 12 creates context information based on the status information acquired by the status acquisition unit 10 (step S2). The notification information creation unit 11 creates notification information based on the status information acquired by the status acquisition unit 10 (step S3). Note that the processing of step S2 and the processing of step S3 may be performed in either order.

[0157] The notification determination unit 13 receives notification information and context information. The notification determination unit 13 selects a fixed value based on the notification information (step S10). Specifically, the notification determination unit 13 selects a fixed value corresponding to the notification information based on the urgency parameter set 2A (urgency data structure). As a result, the notification determination unit 13 determines the selected fixed value as the fixed value of the notification information (step S20).

[0158] The notification determination unit 13 also selects a variable value based on the notification information and the context information (step S30). Specifically, the notification determination unit 13 selects a variable value corresponding to the combination of the notification information and the context information. The notification determination unit 13 then determines the selected variable value as the variable value of the notification information (step S40).

[0159] The notification determination unit 13 may execute the first process, which is the process of steps S10 and S20, and the second process, which is the process of steps S30 and S40, in any order. That is, the notification determination unit 13 may execute either the first process or the second process first, or may execute the first process and the second process simultaneously.

[0160] The notification determination unit 13 determines (determines) the urgency of notification of the notification information based on the notification information and the context information (step S50). Specifically, the notification determination unit 13 determines the urgency based on the determined fixed value and variable value. As described above, there are no particular limitations on the method for determining the urgency, but here we will explain an example of a method of calculating the sum of two values.

[0161] For example, if the fixed value urgency of the notification information is 3.0 and the variable value corresponding to the context information is 2.3, the notification determination unit 13 calculates the final urgency as 3.0 + 2.3 = 5.3.

[0162] The notification determination unit 13 outputs the determined urgency level to the display unit 14. The display unit 14 determines a display method based on the urgency level (determination result) (step S60), and displays the notification information based on the display method (step S70). As a result, the notification information is displayed to the user.

[0163] For example, as described above, when the urgency level is 5.3, if the display unit 14 determines the display method using the determination method using the first relationship in Figure 8, since the urgency level is higher than 5 and equal to or lower than 10, the display unit 14 determines that the display method for the notification information is to be pop-up display.

[0164] The evaluation acquisition unit 15 acquires the user's evaluation of the display result of the notification information (step S80), and outputs it to the notification determination unit 13 as evaluation information.

[0165] The notification determination unit 13 updates the variation value corresponding to the evaluation information based on the evaluation information. Specifically, the notification determination unit 13 updates the variation value selected in step S30 based on the evaluation information. That is, the notification determination unit 13 updates the variation value based on the context information and the evaluation information, thereby changing the urgency of the notification information (step S90). As a result, the evaluation information according to the notification information and the context information is fed back. In this case, the notification determination unit 13 may add the evaluation value of the notification information directly to the variation value, or may determine a value to be added to the variation value in the form of a function determined according to the evaluation value. Furthermore, when updating the variation value, the notification determination unit 13 may update the variation value after reflecting a weight according to the context information in the evaluation value.

[0166] The changed variable value is used in the next urgency determination process. As a result, the next time the notification information is displayed, it will be displayed in a display method that corresponds to the changed variable value. That is, in the next display process, the display unit 14 displays the notification information in a display method that reflects the changed urgency.

[0167] Here, a case will be described in which the evaluation acquisition unit 15 is set to output −0.9 as evaluation information when the “Re-notify Later” button 64 shown in Fig. 12 is pressed. In this case, if the notification determination unit 13 adds the evaluation information directly to the variable value, the aforementioned variable value becomes 2.3−0.9=1.4, and the urgency level at the time of the next display will be 3.0+1.4=4.4. Therefore, if the display unit 14 determines the display method using the determination method using the first relationship in Fig. 8, a banner will be displayed next time.

[0168] For example, if a value six times the evaluation value is reflected in the variation value as a weight according to the context information, the variation value will be 2.3 - (0.9 x 6) = -3.1. In this case, the urgency level for the next display will be 3.0 - 3.1 = -0.1, so if the display unit 14 determines the display method using the first relationship in Figure 8, the next display will be an icon display.

[0169] In this way, the control device 1A updates the variable value and uses the updated variable value when determining the next urgency level, thereby changing the display method to be determined.

[0170] As described above, the control device 1A of the first embodiment calculates the level of urgency based on context information corresponding to the status information of the machine tool and notification information corresponding to the status information, and determines a display method for the notification information based on the calculated level of urgency. Furthermore, the control device 1A updates the variable value of the context information based on the user's evaluation of the display method. The control device 1A then calculates the level of urgency using the updated variable value corresponding to the context information, and determines a display method for the notification information based on the calculated level of urgency. In this way, the control device 1A updates the variable value using the context information and the evaluation information, and changes the level of urgency of the notification information by using the updated variable value. This allows the control device 1A to display appropriate notification information according to the status information of the machine tool. Furthermore, because the control device 1A updates the variable value using the evaluation information, it is possible to calculate the level of urgency taking into account the user's circumstances.

[0171] In this way, the control device 1A changes the urgency of the notification information using context information and evaluation information, so that even when the status information of the machine tool changes, notifications of high urgency for the user can be effectively sent.

[0172] Second Embodiment Next, a second embodiment will be described with reference to Figures 15 to 22. In the second embodiment, the urgency level is managed for each cluster of notification information.

[0173] Fig. 15 is a diagram showing the configuration of a control device according to embodiment 2. Of the components in Fig. 15, those that achieve the same functions as those of the control device 1A according to embodiment 1 shown in Fig. 1 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0174] Like the control device 1A, the control device 1B is a computer that controls a machine tool. In addition to the components of the control device 1A, the control device 1B has a cluster identification unit 17 and a user identification unit 18. That is, the control device 1B has a status acquisition unit 10, a notification information creation unit 11, a context information creation unit 12, a notification determination unit 13, a display unit 14, an evaluation acquisition unit 15, an information storage unit 16, the cluster identification unit 17, and the user identification unit 18.

[0175] [Cluster identification unit 17] Generally, in a mechanism for performing adjustment using feedback data, there is a problem (cold start problem) that the adjustment does not work sufficiently when there is little actual feedback data. In the first embodiment, the control device 1A manages the variation value for each piece of context information and further manages the urgency by individually subdividing each piece of notification information. Therefore, unless the same notification information occurs in the same context, evaluation data will not be collected, and it is expected that it will take time to improve accuracy.

[0176] Therefore, in the control device 1B of embodiment 2, the cluster identification unit 17 clusters the notification information according to similarity and manages the urgency for each cluster to which the notification information belongs (hereinafter referred to as a notification information cluster). That is, the cluster identification unit 17 clusters multiple pieces of notification information according to the similarity of the notification information, identifies the notification information cluster, and manages the urgency of the notification information cluster. As a result, the control device 1B shares evaluation information for notification information belonging to the same notification information cluster within the notification information cluster, and efficiently reflects the evaluation information in the variable value.

[0177] The viewpoint of the similarity of the notification information may be, for example, a viewpoint based on a fixed value of the urgency of the notification information, a viewpoint of the similarity of the notification content itself, etc. Note that these viewpoints are merely examples, and the viewpoint of the similarity of the notification information is not limited to these viewpoints. The cluster identification unit 17 can apply any viewpoint with appropriate modifications as long as it can assign data serving as an index for clustering the notification information to each piece of notification information. Furthermore, the cluster identification unit 17 can perform clustering not only from a single viewpoint but also by combining multiple viewpoints.

[0178] As described above, the fixed value is a value of urgency that notification information is generally expected to have, and is a design value at the design stage of the notification information. Therefore, by clustering from this perspective, the cluster identification unit 17 can treat notification information with similar urgency values ​​assumed at the design stage as notification information belonging to the same cluster.

[0179] Specifically, the cluster identification unit 17 determines the number of clusters to divide, and by equally dividing the range from the maximum to minimum fixed value by the number of clusters, defines the range (range of fixed values) into which notification information cluster each notification information will be clustered.

[0180] Furthermore, the cluster identification unit 17 may apply another clustering method, such as clustering the notification information so that the number of notification information belonging to each cluster is equal, taking into account the distribution of fixed values ​​of all notification information.

[0181] Next, we will explain the similarity of the notification content itself. Each notification information has text information that explains the notification content that has been designed in advance. For example, the cluster identification unit 17 performs clustering using the similarity of words that appear in the text information or the similarity between sentences.

[0182] There are various known methods for calculating the similarity between sentences, but the cluster identification unit 17 may apply any calculation method as long as it takes each character string between the sentences as input and outputs the similarity between the target sentences numerically.

[0183] The cluster identification unit 17 converts the text into a format that can be handled by a computer (for example, vector representation), and calculates the similarity between the text vectors (for example, cosine similarity), thereby calculating the similarity.

[0184] Various methods and models for calculating the similarity between sentences have been researched and published. Many of these technologies are also publicly available for anyone to use. Methods and models that support multiple languages, and methods and models that have been improved to increase the accuracy of similarity calculations, are also being researched and published.

[0185] The cluster identification unit 17 of the second embodiment may use these known techniques to generate a learning model using a unique data set and corpus. The cluster identification unit 17 may also use a publicly available trained model to perform fine tuning or transfer learning and then use the result to calculate the similarity. The cluster identification unit 17 outputs the ID of the notification information cluster to the notification determination unit 13.

[0186] [Clustering] Clustering based on the similarity between sentences will now be described in detail. The cluster identification unit 17 converts each sentence of notification information into a vector representation. At this time, the cluster identification unit 17 first vectorizes each word included in the sentence using a method for obtaining word embedded representations. The cluster identification unit 17 may then perform further calculations on these vectors to obtain embedded sentence representations, which are vector representations of the sentences, or may obtain embedded sentence representations directly from the sentences. Examples of the former include Word2vec, GloVe, and FastText, and examples of the latter include Doc2Vec, BERT (Bidirectional Encoder Representations from Transformers), and USE (Universal Sentence Encoder).

[0187] Here, we will explain a case where the cluster identification unit 17 performs clustering based on the similarity between sentences using Word2vec, a method for obtaining word embeddings. Word2vec is a method for obtaining word embeddings in which words are represented as fixed-length vectors, and there are two types of neural network models: CBOW (Continuous Bag-Of-Words) and skip-gram. Both models are composed of neural networks with only one hidden layer between the input and output layers, and the weights of the hidden layer are adjusted by training them on different tasks, and vectors of word embeddings are obtained as parameters for the weights of this hidden layer.

[0188] CBOW is a model that learns by predicting a word from its surrounding words (i.e., context). CBOW acquires embedded word representations by adjusting the extent and weight of the surrounding words used through learning.

[0189] On the other hand, skip-gram is a model that learns by predicting surrounding words when a certain word is given. Skip-gram acquires word embeddings by learning what words appear around a certain word.

[0190] Both CBOW and skip-gram are supervised learning methods, requiring training data. By using sentences contained in notification information as training data, CBOW and skip-gram are expected to obtain word embeddings based on the relationships between words in the notification information. For example, when CBOW and skip-gram are trained using a general corpus, when the word "block" is given, the predicted surrounding words include words related to architecture, materials, and the uses of rectangular objects. By training using, for example, sentences in notification information from a machine tool control device 1B as training data, CBOW and skip-gram can learn that "block" refers to a line in a machining program and obtain characteristic word embeddings such as "cutting" block and "moving" block.

[0191] In addition, the cluster identification unit 17 may be configured to achieve both high accuracy of distributed representation for general vocabulary using a large corpus and learning of vocabulary relationships that are characteristic of the machine tool field described above, by fine-tuning a trained model trained using a large corpus using sentences from notification information.

[0192] The cluster identification unit 17 performs clustering using a clustering method on the sentence distributed representations of each notification information obtained in this way. The cluster identification unit 17 may use not only the above-mentioned cosine similarity but also known methods such as Jaccard similarity and WMD (Word Mover's Distance) as an evaluation index for this clustering.

[0193] The cluster identification unit 17 may also use TF-IDF (Term Frequency-Inverse Document Frequency) to create a TF-IDF matrix for all notification information sentences and perform clustering.

[0194] TF is an index that represents the frequency of occurrence of a word, and is defined as the value obtained by dividing the frequency of occurrence of a word by the sum of the frequencies of occurrence of all words in each document. If a word frequently appears in a document, the TF value will be large, and therefore TF can numerically express characteristic words that are frequently used in that document. As can be seen from the definition of TF, TF is a value calculated for one word in one document.

[0195] IDF, also known as inverse document frequency, is defined as the logarithm of the total number of documents divided by the number of documents containing a certain word. If a certain word is contained in all documents, the IDF will be 0, and as the number of documents containing the word decreases, the denominator in the logarithm decreases and the IDF value increases. Therefore, IDF can numerically express characteristic words that appear only in a particular document. As can be seen from the definition of IDF, IDF is a value calculated for a single word.

[0196] TF-IDF is calculated by multiplying the TF value and the IDF value. When the cluster identification unit 17 applies TF-IDF, the cluster identification unit 17 calculates the TF-IDF of each document and each word, obtains a matrix having each document and each word as an element, and uses this matrix to calculate the similarity between documents.

[0197] For example, when rows represent documents and columns represent words, the cluster identification unit 17 calculates matrix data by calculating the TF-IDF of each element. This allows the cluster identification unit 17 to calculate the cosine similarity between two documents using the row vectors A and B of the two documents, according to the following formula (1).

[0198]

[0199] The numerator in equation (1) is the dot product of the vectors, and the denominator is the product of the magnitudes of the vectors. Clustering is a data analysis technique that classifies data into several groups and is classified as unsupervised machine learning. Various clustering techniques have been researched and published. The cluster identification unit 17 of embodiment 2 is capable of applying various techniques without any particular limitations to these techniques.

[0200] There are several types of clustering methods, such as hierarchical clustering, in which data is classified hierarchically; non-hierarchical clustering, which does not have a hierarchical structure; hard clustering, in which data always belongs to one group (cluster); and soft clustering, which calculates the probability that data belongs to multiple groups (clusters).

[0201] Hierarchical clustering methods include Ward's method, group average method, shortest distance method, and longest distance method, while non-hierarchical clustering methods include k-means method and its improved methods (k-means++, x-means, etc.).

[0202] Here, the calculation procedure of the k-means method will be described as an example of a non-hierarchical clustering method. When the cluster identification unit 17 uses the k-means method, the cluster identification unit 17 sets the number of clusters to be divided as a hyperparameter. Next, the cluster identification unit 17 sets the initial value of the center of gravity of each cluster to a random value. In this initial state, the cluster identification unit 17 calculates the distance between the center of gravity of each cluster and each piece of data, and clusters each piece of data into the cluster with the closest center of gravity. This is the first clustering.

[0203] The cluster identification unit 17 recalculates the center of gravity of each cluster based on the first clustering result. The cluster identification unit 17 then recalculates the distance between each data item and the recalculated center of gravity of each cluster, and clusters each data item into the cluster with the closest center of gravity. This updates the clustering result.

[0204] The cluster identification unit 17 obtains the clustering result by repeatedly recalculating the center of gravity of each cluster and clustering based on the calculation of the distance between the center of gravity and the data, and repeating the clustering until the center of gravity of each cluster no longer changes due to the recalculation.

[0205] In practice, the cluster identification unit 17 may determine convergence when the change in the center of gravity is equal to or less than a certain threshold, due to the trade-off between calculation time and clustering accuracy. In the k-means method, the number of clusters to be divided is a hyperparameter, as described above, and it is known that the value of this hyperparameter affects the quality of the clustering results. The cluster identification unit 17 uses, for example, the Elbow method as a method for determining the number of clusters. In the Elbow method, the number of clusters is gradually increased, and the point at which the improvement in the index value of the clustering result becomes gradual is adopted as the number of clusters.

[0206] Fig. 16 is a diagram for explaining the Elbow method used by the control device according to the second embodiment. Fig. 16 shows an image of the Elbow method. The horizontal axis of the graph shown in Fig. 16 represents the number of clusters of the notification information, and the vertical axis represents the index value of the clustering result.

[0207] As shown in Fig. 16, it can be seen that the index value becomes gentle when the number of clusters is around 25. When such a trend is obtained, the Elbow method estimates that 25 is the optimal number of clusters.

[0208] The cluster identification unit 17 uses, for example, the intra-cluster sum of squared errors of prediction (SSE) expressed by the following equation (2) as an index value of the clustering result.

[0209]

[0210] In equation (2), k is the number of clusters, C i is the set of data contained in the i-th cluster, and p is C i The index value of the clustering result is the sum of the squares of the distances between the data in each cluster and the center of gravity for all clusters. Note that the cluster identification unit 17 may use a method other than the Elbow method to determine the number of clusters, and may use an index value other than SSE.

[0211] Here, an image of clustering will be explained using an example in which TF-IDF and the k-means algorithm are used in combination. Fig. 17 is a diagram for explaining vector data of TF-IDF used by the control device according to the second embodiment.

[0212] In FIG. 17 , the TF-IDF vector data for each document is represented as a two-dimensional vector for ease of visualization. As described above, each row of the TF-IDF matrix corresponds to one document, and in the second embodiment, each row corresponds to one sentence of notification information. The number of columns of the TF-IDF, i.e., the dimension of the vector, corresponds to the number of words (vocabulary size) used to calculate the TF-IDF, and is therefore typically high-dimensional, such as hundreds or thousands. However, to facilitate a better understanding of clustering, the vector is shown as a two-dimensional vector. Note that the subsequent processing is not limited to two dimensions, and the cluster identification unit 17 can similarly perform clustering on high-dimensional vectors. Each point shown in FIG. 17 corresponds to one notification information, and the cluster identification unit 17 clusters these data groups using the k-means method.

[0213] Fig. 18 is a diagram for explaining clustering by the k-means method used by the control device according to the second embodiment. Fig. 18 shows an image of clustering by the k-means method.

[0214] Each point in Fig. 18 represents the same data group as in Fig. 17, but the clustering results are represented by different types of points in Fig. 18. Also, in Fig. 18, the center of gravity of each cluster is represented by a white dot.

[0215] As described above, the state in which the cluster identification unit 17 determines the number of clusters to be 4 and assigns the initial values ​​of the centers of gravity of the clusters is state G1 (st1).

[0216] Next, the cluster identifying unit 17 performs the first clustering, resulting in clustering of each piece of notification information as shown in state G2 (st2). In state G2, points of the same type belong to the same cluster.

[0217] The cluster identification unit 17 recalculates the center of gravity for each cluster for state G2, thereby moving the center of gravity and changing state G2 to state G3 (st3). After this, the cluster identification unit 17 performs clustering again, updating the clustering result to state G4 (st4). The cluster identification unit 17 repeats the calculations performed in st3 and st4 to change the center of gravity, and obtains the final clustering result when the change in the center of gravity has converged.

[0218] In this way, the cluster identification unit 17 expresses the features of the sentences of each piece of notification information as vectors using TF-IDF, and clusters the point groups represented by each vector using the k-means method. Note that although the example of a two-dimensional vector has been described here, the cluster identification unit 17 can calculate the center of gravity and distance even for higher-dimensional vectors in the same way as for two-dimensional vectors.

[0219] The cluster identification unit 17 assigns identification information such as an ID for identifying the notification information cluster belonging to each notification information to the clustering result. The cluster identification unit 17 sends the notification information cluster (the notification information cluster identification result) with the assigned identification information to the notification determination unit 13. This allows the notification determination unit 13, when selecting one piece of notification information, to select the notification information cluster to which the selected notification information belongs.

[0220] In the second embodiment, the case where the control device 1B uses hard clustering in which one notification information cluster corresponds to one notification information will be described, but the control device 1B may also apply soft clustering. When the notification determination unit 13 selects multiple notification information clusters with non-zero belonging probabilities for one notification information, the notification determination unit 13 may use each probability as a weight by which the fluctuation value of each notification information cluster is multiplied.

[0221] In the second embodiment, for the sake of simplicity, simple non-hierarchical clustering has been described as a premise, but the cluster identification unit 17 can also use hierarchical clustering. In this case, the cluster identification unit 17 performs a modification in which the notification information clusters are staged according to the hierarchical levels, and the notification information clusters are set as a first hierarchical level notification information cluster, a second hierarchical level notification information cluster, etc. Then, the cluster identification unit 17 sets a variable value for each cluster in each hierarchical level, thereby treating the hierarchical clustering as an application of non-hierarchical clustering.

[0222] Furthermore, in the second embodiment, the case has been described in which the control device 1B executes the clustering process for creating notification information clusters from notification information and the process for calculating similarities when clustering, but these processes may be performed in advance by an external computing device such as a personal computer, cloud, or server (not shown). In this case, the external device inputs, as a calculation result, data describing the correspondence relationship as to which notification information cluster each piece of notification information belongs, or logic such as a decision tree, to the control device 1B. This enables the cluster identification unit 17 to identify the notification information cluster.

[0223] For example, the cluster identification unit 17 may be implemented with vector-format data having, as elements, the IDs of notification information clusters corresponding to the IDs of the notification information, or with decision tree logic that derives the notification information clusters from the IDs of the notification information. In this case, when the cluster identification unit 17 receives the IDs of the notification information from an external device, it uses the IDs of the notification information as an index and, based on the vector-format data or the decision tree logic, derives and outputs the IDs of the notification information clusters to which the notification information of the received ID belongs.

[0224] [User Identification Unit 18] The user identification unit 18 identifies the user who uses the machine tool and outputs user information to the notification determination unit 13 according to the identification result. Specifically, the user identification unit 18 compares the user ID input to the control device 1B with registered user information to identify the corresponding user. The control device 1B may manage user information for each user (individual) or for multiple users. When managing user information for multiple users, the control device 1B sets the same user attribute for common or similar users and manages the user information for each user attribute.

[0225] There are no particular limitations on the method by which the control device 1B identifies a user. For example, an identifier such as a user ID that is unique to each user may be registered in advance in the control device 1B. In this case, the user identification unit 18 identifies which user is using the control device 1B based on the user ID and password.

[0226] The control device 1B may identify the user by an identification method using biometric data such as a fingerprint, iris, voiceprint, or face instead of a password. The control device 1B may also authenticate the user using a smartphone, an application, or the like, and acquire the user information of the user as the authentication result. The user may also be identified by combining an external medium storing authentication data, such as an IC (Integrated Circuit) chip or a magnetic card, with an authentication data reader. Furthermore, the control device 1B may enhance security by using multi-factor authentication that combines these identification methods.

[0227] Note that any means may be used to input the above-described user ID to the control device 1B. For example, the control device 1B may accept the user ID and password using an input interface provided in the control device 1B, or may accept the user ID and password by connecting other known input devices or sensors to the control device 1B.

[0228] The control device 1B can determine the display method using the urgency level that is in line with the characteristics of the user by managing the urgency level for each user, rather than by managing the urgency level in user attribute units. For example, the control device 1B manages the urgency level for each user when the number of users who may use the machine tool is less than a specific number.

[0229] When the control device 1B manages the urgency level in units of user attributes, the control device 1B classifies the user attributes according to, for example, whether the user belongs to a machine tool manufacturer or a company or factory that owns the machine tool.

[0230] In addition, the control device 1B may distinguish between user attributes, for example, among users belonging to a machine tool manufacturer, based on whether the user is a serviceman who provides services such as maintenance, inspection, or repair, or a machining engineer who actually uses the machine tool to provide training or adjust its operation.

[0231] In addition, the control device 1B may classify user attributes according to whether the user belongs to a company or factory that owns a machine tool and is a maintenance personnel who similarly performs maintenance, inspection, repair, etc., a user who actually performs processing or production using the machine tool, or a production engineer who performs testing for prototypes or improvements.

[0232] When managing the urgency level on a user attribute basis, the control device 1B may identify the user using the user identification method described above, and then identify the user attributes of the user by referring to a database of user attributes previously assigned to each user. Alternatively, the user attributes may be identified by combining an external medium such as an IC chip or magnetic card on which information on the user attributes is recorded and a reading device that reads the information on the user attributes from the external medium.

[0233] [Context Information Creation Unit 12] Next, a description will be given of the features of the context information creation unit 12 in the second embodiment. The context information creation unit 12 in the second embodiment creates two new types of context information: a short context and a long context.

[0234] As described above, the context information is obtained by performing data processing such as statistical processing on the status information. The context information creation unit 12 grasps the characteristics of the operating status of the machine tool and the operating status of the machine tool based on, for example, the frequency of appearance of the status information over a certain period of time, and creates the grasped characteristics as context information.

[0235] Even with this method of creating context information, the context information creation unit 12 is able to grasp the above-mentioned operating or manipulation status. However, it is anticipated that there will be a demand for even more accurate situation understanding in situations such as those described below.

[0236] The first situation is, for example, a situation in which a machine tool that has been operating unmanned for some time without receiving any user operation generates notification information due to the most recent user operation (for example, some abnormality has occurred). In this case, if the context information creation unit 12 performs statistical data processing, there is a high possibility that it will create context information for unmanned operation, but the situation of the most recent user operation is highly relevant to the notification information. In this case, it cannot be said that appropriate context information has been selected for the notification information.

[0237] The second situation is, for example, the exact opposite of the first situation, in which a user performs an operation or task for a while, then automatic driving starts, and notification information is generated immediately after the start of the automatic driving. In this case, as in the first situation, when the context information creation unit 12 performs statistical data processing, it is highly likely that it will create context information about the setup (user operation or task), but it is highly likely that the content of the most recent automatic driving is highly relevant to the notification information. In this case, as in the first situation, it cannot be said that appropriate context information has been selected.

[0238] To deal with the first or second situation, the context information creating unit 12 in the second embodiment creates two types of context information: a short context that accurately represents the most recent situation, and a long context that accurately represents the long-term average situation. These two types of context information will be described in detail later.

[0239] In the two situations exemplified above, the context information creating unit 12 creates a context corresponding to the context information of embodiment 1 as a long context, and a context corresponding to the most recent situation as a short context, thereby expressing the context by dividing the situations. In this way, the context information creating unit 12 allows the user to accurately grasp the operating status of the machine tool and the operating status of the machine tool when notification information is generated.

[0240] The short context is a context that represents the state immediately before the notification information should be displayed, and is determined using state information for a period (first period) shorter than that of the long context. There are no particular restrictions on the specific period covered by the short context, as long as it is shorter than the period covered by the long context. The period covered by the short context is preferably a relatively short period, such as several tens of minutes to several seconds, but may also be a short period of approximately one hour or less.

[0241] Furthermore, since the operating status and the manipulation status are related to the length of the acquisition cycle for each type of notification information, the period covered by the short context may be set not by time but by a multiple of the acquisition cycle, for example, 10 or 50 cycles of the notification information acquisition cycle.

[0242] The long context is a context that represents a long-term state from a certain point in the past when the machine tool was operating to the point in time when the notification information is displayed, and is determined using state information for a period (second period) longer than that of the short context. There are no particular restrictions on the specific period that is the subject of the long context, as long as it is longer than the period that is the subject of the short context. The period that is the subject of the long context is preferably a period of one week or one month, but may be a period longer than half a day (12 hours) or one day.

[0243] As described above, the long context is determined using state information spanning a long period of time. Therefore, if all data (state information) is retained and the long context is determined, the data size will be large. In this case, an external storage area may be used, but the context information creation unit 12 may determine the long context using a simple method. When determining the long context using a simple method, the context information creation unit 12 may, for example, count only the number of occurrences of each piece of state information and determine the long context based on the ratio of each piece of state information to the sum of all the state information. In this way, by determining the long context using a simple method, the context information creation unit 12 can keep the data size small because it only needs to retain the number of occurrences of each piece of state information.

[0244] Here, a process in which the context information creating unit 12 resets the counter for counting the number of occurrences when the setting period for the long context has expired will be described. When the setting period for the long context has expired, the context information creating unit 12 may reset the counter for counting the number of occurrences by any method. Here, a specific example of a reset method by the context information creating unit 12 will be described.

[0245] When the context information creating unit 12 manages the state information based on the number of occurrences of the state information, the context information creating unit 12 resets the counter for the number of occurrences using the following first reset method or second reset method.

[0246] In the first reset method, the context information creation unit 12 resets each counter of the number of occurrences when the set period has elapsed. This allows the context information creation unit 12 to grasp the situation within the target period of the long context through simple processing. By resetting each counter of the number of occurrences when the set period has elapsed, the context information creation unit 12 does not retain the influence of the previous usage environment, so it is possible to grasp the situation within the target period of the long context even if the usage environment of the machine tool changes significantly.

[0247] In the second reset method, the context information creation unit 12 divides the set period into multiple intervals (e.g., three intervals) and divides each occurrence count counter into the same number of intervals. In this case, the context information creation unit 12 resets only the counter for the oldest interval among the divided intervals. As a result, the context information creation unit 12 retains data (occurrence counts) for the remaining intervals (e.g., two intervals), so that even before and after resetting the counters, it is possible to reduce changes in the state information and mitigate degradation in the accuracy of the long context.

[0248] When managing state information based on the number of occurrences of state information, the context information creation unit 12 may apply either the first reset method or the second reset method, or may apply both. When using both the first reset method and the second reset method, the context information creation unit 12 can select an appropriate reset method according to the situation by selecting or changing which counter to reset when the set period has elapsed.

[0249] Furthermore, when the context information creating unit 12 holds all data and determines a long context, it clears all data in the same manner as in the first reset method, for example.

[0250] The control device 1B may also store all data together with time-series information. In this case, the context information creating unit 12 can update data with high accuracy by deleting or overwriting data in order from oldest to newest.

[0251] Furthermore, the context information creation unit 12 may reset the counter in response to a trigger other than the set period. For example, the context information creation unit 12 resets the counter when the power to the machine tool is turned off. In this way, in a usage environment where the machine tool operates 24 hours a day and the power is turned off only at special times such as during maintenance or when the machining content is changed, an event that could be a change in the usage situation will coincide with the reset trigger, and the control device 1B can appropriately manage the status information.

[0252] Alternatively, the control device 1B may prompt the user to explicitly execute a reset operation. In this case, the control device 1B resets the counter when the user executes a reset operation.

[0253] The content of the two types of context information described above (long context and short context) is not significantly different from the context information described in the first embodiment. That is, status information such as "automatic operation in progress" can be either a long context or a short context. Furthermore, the control device 1B manages the two types of context information using an identifier such as a unique number, similar to the context information in the first embodiment. Furthermore, the operating status of the machine tool in the two types of context information includes an operation context and an operation context, similar to the context information in the first embodiment.

[0254] The context information creation unit 12 may treat the two types of context information as separate data and output them as a set of data included in the context information. Alternatively, the context information creation unit 12 may output the two types of context information as a single piece of context information determined by a combination of the two pieces of context information. In this case, the context information creation unit 12 may represent the two types of context information in a matrix data format, for example, by taking the IDs of the long contexts in the rows and the IDs of the short contexts in the columns. That is, the context information creation unit 12 may create a data structure for the context information in which identifiers of the context information represented by a combination of two pieces of context information are stored as elements of a corresponding matrix.

[0255] [Operation of notification determination unit 13] In addition to the operation in the above-described embodiment 1, the notification determination unit 13 in embodiment 2 further determines the level of urgency by using at least one of the identified user identification information, the identified notification information cluster (the identification result of the notification information cluster), the long context, the short context, and the state information.

[0256] The notification determination unit 13 can apply these pieces of information used when determining the level of urgency independently of each other, and there is no mutual dependency between them. Therefore, the notification determination unit 13 may selectively use multiple types of information in combination as information used when determining the level of urgency, for example, by using user identification information and notification information clusters, or may apply all types of information. Furthermore, even if the notification determination unit 13 applies the above-mentioned information alone as information used when determining the level of urgency, it can obtain an effect according to the applied information.

[0257] Here, a case where the notification determination unit 13 determines the urgency level using the notification information cluster will be described. Fig. 19 is a diagram for explaining an example of an urgency parameter set when the control device according to the second embodiment determines the urgency level using the notification information cluster. Note that, among the components in Fig. 19, components that achieve the same function as the urgency parameter set 2A of the first embodiment shown in Fig. 2 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0258] The urgency parameter set 2B1 shown in Fig. 19 includes a notification information cluster. In Fig. 19, notification information with an ID of "001" and notification information with an ID of "002" are classified into the notification information cluster with an ID of "001". Furthermore, notification information with an ID of "xxx" is classified into the notification information cluster with an ID of "xxx".

[0259] Various context information such as contexts Ax, Bx, and Cx are associated with the notification information cluster with ID "001." Similarly, various context information such as contexts Ax, Bx, and Cx are associated with the notification information with ID "xxx."

[0260] When the notification determination unit 13 determines the urgency level using the notification information cluster, it is desirable to manage at least the variation value of the urgency level as a separate parameter for each notification information cluster, as shown in Fig. 19. This allows evaluation information for multiple notification information items to be reflected in the variation value of the notification information cluster to which these notification information items belong. Therefore, the control device 1B can update the urgency level based on the evaluations for notification information items belonging to the same notification information cluster, even before evaluations for individual notification information items are accumulated.

[0261] Furthermore, when determining the urgency level using the long context and the short context, there are, for example, a first management method and a second management method. In the first management method, it is desirable that the notification determination unit 13 further divides the context information, which has been managed as a separate parameter for each piece of context information, into two and manages them. In this case, since the variation value is managed as a separate parameter for each piece of context information, two parameters, one for the variation value for each long context and one for the variation value for each short context, exist for each piece of notification information.

[0262] Fig. 20 is a diagram for explaining an example of an urgency parameter set when the control device according to the second embodiment determines the urgency using a long context and a short context. Note that, among the components in Fig. 20, components that achieve the same functions as those in the urgency parameter set 2A of the first embodiment shown in Fig. 2 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0263] An urgency parameter set 2B2 shown in FIG. 20 is an urgency parameter set when the control device 1B manages urgency using a long context and a short context.

[0264] 20 shows a case where notification information with an ID of "001" includes a long context and a short context. The long context included in the notification information with an ID of "001" is associated with various context information such as long contexts LA, LB, and LC. Similarly, the short context included in the notification information with an ID of "001" is associated with various context information such as short contexts SA, SB, and SC. The short contexts SA, SB, and SC are context information corresponding to the long contexts LA, LB, and LC, respectively.

[0265] When determining one variation value from two variation values ​​consisting of a variation value (parameter) of a long context and a variation value of a short context, the notification determination unit 13 may simply determine the variation value to be the sum of the two variation values.

[0266] Furthermore, the notification determination unit 13 may determine the output value of a function expressed in the form of a mathematical formula with two variable values ​​as variables as the variable value. That is, the notification determination unit 13 may determine the value obtained by substituting two variable values, the variable value of the long context and the variable value of the short context, into the function expressed in the form of a mathematical formula as the variable value.

[0267] In a second urgency management method, the notification determination unit 13 manages the variation value for each combination of a long context and a short context as a separate parameter. For example, the notification determination unit 13 manages the combination of the variation value of the long context LA and the variation value of the short context SA as a separate parameter XA. This allows the notification determination unit 13 to reflect an evaluation based on the interrelationship between the long context and the short context in the variation value.

[0268] The notification determination unit 13 may also manage the urgency level using user identification information. In this case, it is desirable for the notification determination unit 13 to manage the urgency level individually for each user. Figure 21 is a diagram for explaining the process in which the control device according to the second embodiment manages the urgency level for each user using user identification information.

[0269] When the notification determination unit 13 uses user identification information, information associating the user identification information with all of the urgency parameter sets is stored in the information storage unit 16. Fig. 21 shows a case in which user identification information U1 is associated with urgency parameter set P1, and user identification information U2 is associated with urgency parameter set P2. The urgency parameter sets P1 and P2 are, for example, urgency parameter sets 2A, 2B1, and 2B2.

[0270] When the notification determination unit 13 uses user identification information such as user identification information U1 and U2, it is desirable that the entire urgency parameter set P1 and P2 be managed individually for each identified user using the user identification information U1 and U2, as shown in Figure 21.

[0271] As described above, user identification information may be managed for each user or for each user attribute. When user identification information is managed for each user, detailed management is possible. When user identification information is managed for each user attribute, the number of parameters can be reduced. Therefore, it is desirable to determine how to use these in light of the usage situation in which the machine tool is placed. For example, in cases where the number of users is limited, such as in a small factory or a highly specialized machine tool (machine), user identification information is managed for each user. On the other hand, in cases where the number of users is wide, such as in a large factory or a general-purpose machine tool, user identification information is managed for each user attribute.

[0272] 22 is a flowchart showing the procedure of processing executed by the control device according to the second embodiment. Here, the processing executed by the control device 1B to determine the urgency level, determine the display method, and update the urgency level based on the evaluation will be described.

[0273] In the second embodiment, similarly to the first embodiment, the process of acquiring status information, the process of creating context information, and the process of creating notification information are executed. The user identification unit 18 determines whether or not there is user information (step S110). That is, the user identification unit 18 determines whether or not there is user information stored in the information storage unit 16. If there is user information (Yes in step S110), the user identification unit 18 identifies the user who uses the machine tool based on the user information, and outputs the user information to the notification determination unit 13 according to the identification result. Specifically, the user identification unit 18 compares the user ID input to the control device 1B with registered user information to identify the corresponding user, and outputs the user information corresponding to the user to the notification determination unit 13.

[0274] The notification determination unit 13 selects an urgency parameter set corresponding to the user based on the user information (step S120), after which the cluster identification unit 17 executes the process of step S130.

[0275] If there is no user information (No in step S110), the control device 1B performs subsequent processing using a default urgency parameter set. That is, if the control device 1B is configured not to have a user identification unit 18 or if the user is an unregistered user, the control device 1B performs subsequent processing using a default urgency parameter set that is not associated with user identification information.

[0276] After the notification determination unit 13 selects the urgency parameter set corresponding to the user, or if there is no user information, the cluster identification unit 17 determines whether or not there is a notification information cluster (step S130). That is, the cluster identification unit 17 determines whether or not a notification information cluster is stored in the information storage unit 16.

[0277] If there is no notification information cluster (step S130, No), the notification determination unit 13 selects a fixed value corresponding to the notification information based on the urgency parameter set (step S140), and then determines the selected fixed value as the fixed value of the notification information (step S150).

[0278] The notification determination unit 13 determines whether or not the context information created by the context information creation unit 12 includes a plurality of pieces of context information (step S160). That is, the notification determination unit 13 determines whether or not the context information created by the context information creation unit 12 includes a long context and a short context.

[0279] If there are not multiple pieces of context information (No at step S160), the notification determination unit 13 selects a variable value based on the notification information and one piece of context information (step S170).

[0280] On the other hand, if there are multiple pieces of context information (Yes at step S160), the notification determination unit 13 selects a variable value based on the notification information and the multiple pieces of context information, that is, LS context information (step S180). The LS context information is context information having a long context and a short context.

[0281] By the process of step S170 or step S180, the notification determination unit 13 determines the selected variable value as the variable value of the notification information (step S190).

[0282] If there is a notification information cluster (step S130, Yes), the notification determination unit 13 selects a fixed value corresponding to the notification information cluster based on the urgency parameter set (step S200), and then determines the selected fixed value as the fixed value of the notification information (step S210).

[0283] The notification determination unit 13 determines whether the context information created by the context information creation unit 12 includes a plurality of pieces of context information (step S220). That is, the notification determination unit 13 determines whether the context information created by the context information creation unit 12 includes a long context and a short context.

[0284] If there are not multiple pieces of context information (No at step S220), the notification determination unit 13 selects a variation value based on the notification information cluster and one piece of context information (step S230).

[0285] On the other hand, if there are a plurality of pieces of context information (Yes at step S220), the notification determination unit 13 selects a variation value based on the notification information cluster and the LS context information, which is a plurality of pieces of context information (step S240).

[0286] Through the process of step S230 or step S240, the notification determination unit 13 determines the selected variable value as the variable value of the notification information (step S190).

[0287] The control device 1B may execute the processes of steps S140 and S150 and the processes of steps S160 to S180 in any order. The control device 1B may execute the processes of steps S200 and S210 and the processes of steps S220 to S240 in any order.

[0288] Thereafter, the control device 1B executes steps S250 to S290, which are the same as steps S50 to S90 described in FIG. 14 of the first embodiment. That is, the notification determination unit 13 determines the urgency level based on the notification information and context information (step S250). Specifically, the notification determination unit 13 determines the urgency level based on the determined fixed value and variable value. The display unit 14 determines a display method based on the urgency level (step S260) and displays the notification information based on the display method (step S270). As a result, the notification information is displayed to the user.

[0289] The evaluation acquisition unit 15 acquires the user's evaluation of the display result of the notification information (step S280), and the notification determination unit 13 updates the variable value corresponding to the evaluation information based on the evaluation information. Specifically, the notification determination unit 13 updates the variable value selected in any of steps S170, S180, S230, and S240 based on the evaluation information. That is, the notification determination unit 13 updates the variable value based on the context information and the evaluation information, thereby changing the urgency of the notification information (step S290).

[0290] When the notification determination unit 13 selects a variable value based on the notification information and the context information (when step S170 is executed), the notification determination unit 13 updates the variable value using the context information and the evaluation information, thereby changing the urgency of the notification information. As a result, evaluation information according to the notification information and the context information is fed back.

[0291] Furthermore, when the notification determination unit 13 selects a variable value based on the notification information and the LS context information (when step S180 is executed), it updates the variable value using the LS context information and the evaluation information, thereby changing the urgency of the notification information. As a result, evaluation information according to the notification information and the LS context information is fed back.

[0292] Furthermore, when the notification determination unit 13 selects a variable value based on the notification information cluster and the context information (when step S230 is executed), it updates the variable value using the notification information cluster, the context information, and the evaluation information, thereby changing the urgency of the notification information. As a result, evaluation information according to the notification information cluster and the context information is fed back.

[0293] Furthermore, when the notification determination unit 13 selects a variable value based on the notification information cluster and the LS context information (when step S240 is executed), the notification determination unit 13 updates the variable value using the notification information cluster, the LS context information, and the evaluation information, thereby changing the urgency of the notification information. As a result, evaluation information according to the notification information cluster and the LS context information is fed back.

[0294] In this way, the control device 1B of the second embodiment uses the notification information cluster to integrate the evaluations of similar notification information and update the urgency level, thereby efficiently optimizing the display method.

[0295] Furthermore, since the control device 1B uses a long context and a short context, it is possible to grasp the usage and operation status of the machine tool in detail, and to optimize the display method with high precision.

[0296] Furthermore, since the control device 1B uses user information, it is possible to manage the urgency level for each user of the machine tool, and the display method can be optimized with high accuracy.

[0297] These effects are exerted independently. That is, the control device 1B can exert its effects alone without other elements, and can exert each effect without impairing the other effects even when combined with other elements. For example, if the control device 1B does not have the user identification unit 18 and has only the cluster identification unit 17, the effect of efficiently optimizing the display method by the cluster identification unit 17 can be obtained, and if both are provided, the effect of highly accurately optimizing the display method by the user identification unit 18 can also be obtained.

[0298] Third Embodiment Next, a third embodiment will be described with reference to Fig. 23. In the third embodiment, a user's evaluation is inferred based on state information.

[0299] Fig. 23 is a diagram showing the configuration of a control device according to embodiment 3. Among the components in Fig. 23, components that achieve the same functions as those in the control device 1B according to embodiment 2 shown in Fig. 15 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0300] Like the control device 1B, the control device 1C is a computer that controls a machine tool. In addition to the components of the control device 1B, the control device 1C has an evaluation inference unit 19. That is, the control device 1C has a state acquisition unit 10, a notification information creation unit 11, a context information creation unit 12, a notification determination unit 13, a display unit 14, an evaluation acquisition unit 15, an information storage unit 16, a cluster identification unit 17, a user identification unit 18, and an evaluation inference unit 19.

[0301] [Evaluation inference unit 19] The evaluation inference unit 19 extracts state feature quantities from the state information acquired by the state acquisition unit 10 and stored in the information storage unit 16, and outputs an evaluation inference value based on these state feature quantities. The state feature quantities are feature quantities of the state information derived from the state information. The state feature quantities are feature quantities that can be captured from the state information itself, the trend of the state information, changes in the state information, etc. Specifically, the state feature quantities are characteristic state information, characteristics of the time-series trend of the state information, or characteristics of changes in the state information, and are information that expresses these in the form of feature quantities. The state feature quantities may be information in any format as long as they are information created by analyzing state information other than the user's direct evaluation behavior and capturing the above-mentioned features.

[0302] The evaluation inference value is an inferred value of the user's evaluation of the notification information, inferred based on the state feature amount. The evaluation inference value is not limited to a value that is assumed to be used as the value of the evaluation information as is by the evaluation acquisition unit 15, but may also be a value that is a part of a parameter of a calculation formula used by the evaluation acquisition unit 15 to calculate the value of the evaluation information.

[0303] In the first embodiment, the evaluation acquisition unit 15 was described as acquiring evaluation information by a direct evaluation action by the user, such as pressing the "useful" button 61 or the "not useful" button 62. However, the user does not necessarily take such an action every time. Therefore, the evaluation inference unit 19 in the third embodiment acquires an evaluation by inferring what evaluation the user has made of the notification information, without relying on such a direct evaluation action.

[0304] For example, consider a case where the operating state of the status information is an abnormal state. In this case, notification information for the abnormal state that has occurred is displayed. The evaluation inference unit 19 infers that the shorter the time until the abnormal state corresponding to the displayed notification information is resolved, the more useful the notification information was to the user. On the other hand, the evaluation inference unit 19 infers that the longer the time until the abnormal state corresponding to the displayed notification information is resolved, the less useful the notification information was to the user. In this way, the time from when the notification information is displayed until the machine tool enters a specific state (when the abnormal state is resolved) is an example of a status feature.

[0305] Furthermore, the evaluation inference unit 19 may use the frequency of changes in the screen state within a certain period of time, which is included in the state information, as the state feature. For example, the evaluation inference unit 19 compares the frequency of changes in the screen state before and after the notification information is displayed. In this case, if the frequency of changes after the notification information is displayed is higher than before the notification information is displayed, the user may be checking the notification information or performing operations based on the content of the notification information, and the evaluation inference unit 19 infers that the notification information is being utilized. On the other hand, if there is no significant change in the frequency of changes in the screen state before and after the notification information is displayed, the user may be continuing the same task without being affected by the notification information, and the evaluation inference unit 19 infers that the notification information is not being utilized.

[0306] Furthermore, the evaluation inference unit 19 may use information indicating whether a specific operation has been performed after the notification information has been displayed as the state feature. Specifically, the evaluation inference unit 19 may use information such as whether an operation described in the notification information has been performed or whether the information described in the notification information has been confirmed, in combination with the content included in the notification information, as the state feature. That is, the evaluation inference unit 19 may use a combination of the content included in the notification information and information indicating whether the operation described in the notification information has been performed as the state feature. Furthermore, the evaluation inference unit 19 may use a combination of the content included in the notification information and information indicating whether the information described in the notification information has been confirmed as the state feature. In these cases, the evaluation inference unit 19 infers a positive evaluation inference value when an action based on the operation or information described in the notification information has been performed.

[0307] The state feature includes at least one of the information described above: the time from when the notification information is displayed until the machine tool reaches a specific state, the frequency of changes in the screen state before and after the notification information is displayed, and whether or not a specific operation is performed after the notification information is displayed.

[0308] The evaluation acquisition unit 15 of the third embodiment may output evaluation information based on the evaluation behavior of the user, or may output evaluation information using an evaluation inference value inferred by the evaluation inference unit 19 .

[0309] As described above, the evaluation acquisition unit 15 may output the value of the evaluation inference value as evaluation information as is. Alternatively, the evaluation acquisition unit 15 may express the relationship for calculating the evaluation information in the form of a mathematical formula or a decision tree, and use the evaluation inference value as one parameter of these to determine the evaluation information and output it to the notification determination unit 13. After the evaluation information is output as described above, the notification determination unit 13 updates the variable value to an appropriate value based on the evaluation information.

[0310] In embodiment 3, the case where the control device 1C has a cluster identification unit 17 and a user identification unit 18 has been described, but the control device 1C does not have to have at least one of the cluster identification unit 17 and the user identification unit 18.

[0311] In this way, the control device 1C of embodiment 3 acquires information about the user's operation or the state of the machine tool from the status information and infers evaluation information based on the status information, even when the user does not perform a direct evaluation action. This allows the control device 1C to acquire evaluation information even when the user does not perform a direct evaluation action, and to appropriately update the display method of notification information based on the evaluation information.

[0312] Fourth Embodiment Next, a fourth embodiment will be described with reference to Fig. 24. In the fourth embodiment, it is determined whether or not an abnormality or warning has occurred based on the status information, and if an abnormality or warning has occurred, the content of the measures to be taken to address the abnormality or warning is included in the notification information and displayed.

[0313] Fig. 24 is a diagram showing the configuration of a control device according to embodiment 4. Among the components in Fig. 24, components that achieve the same functions as those of the control device 1C according to embodiment 3 shown in Fig. 23 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0314] Like the control device 1C, the control device 1D is a computer that controls a machine tool. The control device 1D has a notification information creation unit 70 instead of the notification information creation unit 11. That is, the control device 1C has a status acquisition unit 10, a notification information creation unit 70, a context information creation unit 12, a notification determination unit 13, a display unit 14, an evaluation acquisition unit 15, an information storage unit 16, a cluster identification unit 17, a user identification unit 18, and an evaluation inference unit 19.

[0315] The notification information creation unit 70 of the fourth embodiment includes an abnormality determination unit 71 and a treatment information creation unit 72. Similar to the notification information creation unit 11, the notification information creation unit 70 creates notification information based on the status information acquired by the status acquisition unit 10.

[0316] The abnormality determination unit 71 determines whether or not a warning associated with an abnormality or an unauthorized operation has occurred. If the abnormality determination unit 71 determines that an abnormality has occurred, information about the abnormality is included in the notification information as abnormality information. If the abnormality determination unit 71 determines that a warning associated with an unauthorized operation has occurred, information about the warning is included in the notification information as warning information. Specific examples of the abnormality information and warning information in the notification information have been described in embodiment 1, and therefore will not be described again.

[0317] The abnormality determination unit 71 determines whether an abnormality or warning has occurred in the machine tool based on the change in the status information, and outputs the determination result to the treatment information creation unit 72. The abnormality determination unit 71 also outputs the determination result to the treatment information creation unit 72, including content information indicating the content of the abnormality or warning (information related to the abnormal state).

[0318] The method of determining an abnormality or warning by the abnormality determination unit 71 may be any method based on a change in state information. That is, the method of determining an abnormality in the fourth embodiment is merely an example, and is not intended to limit the fourth embodiment, and various modifications are possible.

[0319] For example, list data of status information for determining whether an abnormality or warning has occurred is prepared in advance in the software of the control device 1D. In this case, the abnormality determination unit 71 determines whether an abnormality or warning has occurred based on whether the status information in the list data has changed from a non-occurrence state to an occurrence state. In this case, the status information in the list data may be, for example, the above-mentioned information on improper operation, information on hardware abnormalities, information on software abnormalities, information on the status of peripheral devices, information on the status of the system, etc.

[0320] The abnormality determination unit 71 may also perform statistical or time-series analysis of the status information. In this case, the abnormality determination unit 71 determines whether or not there is an abnormality or a warning based on whether the result of the statistical or time-series analysis is above or below a predetermined threshold. The abnormality determination unit 71 may also determine whether or not there is an abnormality or a warning based on a comparison between the result of the time-series analysis and a predefined pattern of time-series change. The status information in this case may be, for example, expiration information, capacity information, etc.

[0321] When the treatment information creation unit 72 receives a judgment result indicating that an abnormality or warning has occurred from the abnormality judgment unit 71, it creates treatment information according to the content of the abnormality or warning that has been determined to have occurred, based on content information indicating the content of the abnormality or warning contained in the judgment result.

[0322] The treatment information includes at least the details to be taken to resolve the abnormality or warning that has occurred, i.e., the treatment details. The treatment details do not need to be single, and may be multiple. For example, if multiple causes of the abnormality or warning are considered, there may be multiple candidate treatment details, and the treatment information creation unit 72 may create the treatment information by listing multiple candidates.

[0323] The treatment information creating unit 72 creates treatment information indicating treatment details using, for example, information (treatment correspondence information) in which possible abnormalities or warnings are associated with candidate treatment details. In this case, the treatment information creating unit 72 creates treatment information corresponding to the determination result based on the treatment correspondence information and the determination result from the abnormality determining unit 71. In this way, by previously setting candidate treatment details in association with each possible abnormality or warning, the treatment information creating unit 72 can create treatment information corresponding to the determination result from the abnormality determining unit 71.

[0324] Specifically, an identifier (ID), such as a unique number that can individually identify an abnormality or warning, is assigned to each abnormality or warning. Furthermore, an identifier, such as a unique number that can individually identify each treatment content, is assigned to each treatment content. A storage device, such as the information storage unit 16, stores a list in which the identifiers of abnormalities or warnings are associated with the identifiers of treatment content. The treatment information creation unit 72 references this list, extracts treatment content corresponding to the content information of the determination result from this list, and creates treatment information indicating the treatment content. The treatment information creation unit 72 outputs the treatment information to the notification determination unit 13. The notification determination unit 13 then creates notification information using the treatment information. The notification determination unit 13 displays the notification information including the treatment information.

[0325] In addition, in embodiment 4, the case where the control device 1D has a cluster identification unit 17, a user identification unit 18, and an evaluation inference unit 19 has been described, but the control device 1D does not have to have at least one of the cluster identification unit 17, the user identification unit 18, and the evaluation inference unit 19.

[0326] In this way, the control device 1D of the fourth embodiment determines whether or not there is an abnormality or warning based on the status information of the machine tool, and displays the abnormality or warning information and the details of the measures to be taken in response to the abnormality or warning in the notification information. This allows the control device 1D to provide the user with detailed information necessary for appropriate use of the machine tool.

[0327] Embodiment 5 In embodiment 5, a modified example of embodiments 1 to 4 will be described. In embodiments 1 to 4, when the machine tool and control device 1A to 1D are used, the urgency fluctuation value is changed from moment to moment, thereby changing the display method of notification information to one that is appropriate for the user.

[0328] However, there may be cases where a constantly changing urgency variation value creates an uncomfortable operating environment for the user. For example, consider a case where the display method has been adjusted to an optimal state, the operating conditions change temporarily, and then the operating conditions return to the previous state. In this case, the urgency variation value is adjusted to reflect the temporarily changed operating conditions, which may cause inconvenience in that the display method needs to be adjusted again when the operating conditions return to the previous state. In such cases, the inconvenience of the control devices 1A to 1D can be eliminated by the following three modifications, first to third.

[0329] [First Modification] The control devices 1A to 1D eliminate the inconvenience of unnecessary variable value adjustment by stopping the update of the display method. When a user operates or sets the control device 1A to 1D to stop updating the display method, the control device 1A to 1D notifies the notification determination unit 13 that the update of the variable value has been stopped.

[0330] This allows the control devices 1A to 1D to fix the display method without reflecting the evaluation performed while the update of the variable values ​​was stopped. For example, when the control devices 1A to 1D receive an operation or setting to stop updating from the user while the display method has been adjusted to an appropriate value, they notify the notification determination unit 13 that the update of the variable values ​​has been stopped.

[0331] The control devices 1A to 1D may notify the notification determination unit 13 of the update stop immediately after receiving the operation or setting to stop updating, or may notify the notification determination unit 13 of the update stop when the usage situation changes temporarily. After the usage situation returns to normal, the control devices 1A to 1D notify the notification determination unit 13 of the request to resume updating the display method, and the notification determination unit 13 resumes updating the variable values. This allows the user to use the machine tool while maintaining the display method, even if the usage situation changes temporarily when the display method is just right for the user.

[0332] [Second Modification] The control devices 1A to 1D eliminate the inconvenience of unnecessary variable value adjustment by saving the adjusted urgency parameter set. The control devices 1A to 1D save the currently applied urgency parameter set in a storage area accessible from the control devices 1A to 1D (for example, the information storage unit 16) in an identifiable manner, for example by giving it a name.

[0333] The control devices 1A to 1D store an urgency parameter set when instructed by a user or at specific intervals. The control devices 1A to 1D may store multiple urgency parameter sets. In this case, the control devices 1A to 1D may delete the oldest stored urgency parameter set and store the latest urgency parameter set.

[0334] When the control devices 1A to 1D receive a request from a user to reproduce an urgency parameter set, they read out the urgency parameter set specified in the reproduction request from a storage area. The control devices 1A to 1D can reproduce the saved display method by overwriting the current urgency parameter set with the read urgency parameter set.

[0335] With this method, the control devices 1A to 1D can prepare an urgency parameter set for each use environment or user, and by switching the urgency parameter set each time, can always display notification information in an appropriate display method.

[0336] [Third Modification] The control devices 1A to 1D eliminate the inconvenience of unnecessary adjustment of variable values ​​by storing urgency parameter sets associated with characteristic information of the usage environment. The control devices 1A to 1D store information (urgency information) in a storage area such as the information storage unit 16, in which an identifier such as a number assigned to each urgency parameter set so as to be identifiable is associated with characteristic information characterizing the usage environment of the machine tool. In this case, the control devices 1A to 1D extract the characteristic information of the usage environment from the status information. The control devices 1A to 1D then extract, from the urgency information, an urgency parameter set corresponding to the characteristic information extracted from the status information, and apply the extracted urgency parameter set, thereby automatically switching the urgency parameter set.

[0337] Examples of characteristic information that characterizes the usage environment include a machine number that identifies the machine tool, a product number that identifies the control devices 1A to 1D, a program name of a machining program included in the status information, and an M-code command number.

[0338] In this way, when the control devices 1A to 1D of embodiment 5 receive an operation or setting to stop updating from the user, they stop updating the display method, thereby maintaining the current display method and displaying appropriate notification information.

[0339] Furthermore, by storing the adjusted urgency parameter set, the control devices 1A to 1D can reproduce the urgency parameter set, and can display appropriate notification information.

[0340] In addition, by storing an urgency parameter set associated with characteristic information of the usage environment, the control devices 1A to 1D can automatically set an appropriate urgency level in response to changes in the usage environment, and can display appropriate notification information.

[0341] Here, the hardware configuration of the control devices 1A to 1D will be described. Note that the control devices 1A to 1D have similar hardware configurations, so the hardware configuration of the control device 1A will be described here.

[0342] 25 is a diagram illustrating an example of a hardware configuration for implementing the control device according to the first embodiment. The control device 1A can be implemented by an input device 300, a processor 100, a memory 200, and a display device 400. Examples of the processor 100 include a CPU (Central Processing Unit, also referred to as a central processing unit, processing device, arithmetic unit, microprocessor, microcomputer, or DSP (Digital Signal Processor)) or a system LSI (Large Scale Integration). Examples of the memory 200 include a RAM (Random Access Memory) and a ROM (Read Only Memory).

[0343] The control device 1A is realized by the processor 100 reading and executing a control calculation program (not shown) that is executable by a computer and that is for executing the operation of the control device 1A and that is stored in the memory 200. The control calculation program, which is a program for executing the operation of the control device 1A, can also be said to be a program that causes a computer to execute the procedure or method of the control device 1A.

[0344] The control operation program executed by the processor 100 has a modular configuration including the control device 1A, and these components are loaded onto the main storage device and generated on the main storage device. Specifically, the control operation program executed by the processor 100 has a modular configuration including a status acquisition unit 10, a notification information creation unit 11, a context information creation unit 12, a notification determination unit 13, and an evaluation acquisition unit 15, and these components are loaded onto the main storage device and generated on the main storage device.

[0345] The input device 300 accepts information set by the user, status information, etc., and sends it to the processor 100 or the memory 200. The memory 200 stores a control calculation program (not shown), an urgency parameter set 2A, status information, etc. The memory 200 is also used as a shared area that is a temporary memory when the processor 100 executes various processes. The display device 400 is a liquid crystal monitor or the like, and the processes executed by the display device 400 correspond to the processes executed by the display unit 14.

[0346] The control calculation program may be provided as a computer program product stored in a computer-readable storage medium as an installable or executable file. The control calculation program may also be provided to the control device 1A via a network such as the Internet. Note that some of the functions of the control device 1A may be realized by dedicated hardware such as a dedicated circuit, and some may be realized by software or firmware.

[0347] However, because machine tools are used for a wide variety of processes in a wide range of fields in the manufacturing industry, the way they are used varies greatly depending on the environment in which they are used and the user. Furthermore, even for the same machine tool, the way it is used can vary greatly depending on changes in the operating conditions or the processing content.

[0348] For example, machine tools may be used in automated factories that operate continuously 24 hours a day, or may be used temporarily when necessary, with a lower operating rate than usual. In the former case, it is rare for a user to be in charge of one machine tool all the time, and it is expected that one person will manage multiple machines, or that the machine will be largely unmanned. On the other hand, in the latter case, it is expected that a user will operate one machine tool intensively.

[0349] In addition, machine tools may process large metal objects, or may process very small pieces of resin or wood. When processing large materials, the processing time per material is longer, and when processing small materials, the processing time per material is shorter.

[0350] In the case of long-term machining, if there is a problem along the way, it will have a major impact on the machining, so in some cases, the user wants to carefully check the machining process, such as when machining the first product, or in other cases, because the machining is already proven, and want to leave the machining process until it is complete. Also, in the case of short-term machining, the user may wait for the machining to finish soon, or may want to leave the machining process until the planned number of pieces is completed because they are machining the same number of pieces in succession. Also, depending on the material or machining content, the user may need to perform manual work such as removing chips during machining, and the user may not be able to leave the machining area.

[0351] As described above, it is expected that machine tools will be placed in a wide variety of situations, and therefore there is a demand for a technology that notifies the user of information required for use of the machine tool in an appropriate manner depending on the situation. The control devices 1A to 1D of the first to fifth embodiments change the urgency of notification information using context information and evaluation information, and therefore can display notification information in an appropriate display method depending on the usage situation of the machine tool or the user.

[0352] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or different embodiments may be combined with each other. It is also possible to omit or modify parts of the configurations as long as they do not deviate from the gist of the invention.

[0353] DESCRIPTION OF SYMBOLS 1A to 1D Control device, 2A, 2B1, 2B2, P1, P2 Urgency parameter set, 2a Operation screen, 2b Setup screen, 10 Status acquisition unit, 11 Notification information creation unit, 12 Context information creation unit, 13 Notification determination unit, 14 Display unit, 15 Evaluation acquisition unit, 16 Information storage unit, 17 Cluster identification unit, 18 User identification unit, 19 Evaluation inference unit, 31a, 31b First information display area, 32a, 32b Main screen transition button display area, 33a, 33b Coordinate value display area, 34a Machining program display area, 35a, 35b Second information display area, 36a Control mode display area, 37a, 37b Third information display area, 38a, 38b Main screen menu button display area, 39b Parameter display area, 40b Input data display area, 41a, 41b Left scroll buttons, 42a, 42b, right scroll buttons, 43a, 43b, 44a, 44b, 61 to 65, buttons, 45b, vertical scroll bar, 51, 52, 54 to 56, notification information, 53, icon, 70, notification information creation unit, 71, abnormality determination unit, 72, treatment information creation unit, 100, processor, 200, memory, 300, input device, 400, display device, U1, U2, user identification information, X1, first range, X2, second range, X3, third range, X4, fourth range, sC1 to sC9, screens.

Claims

1. A control device comprising: a status acquisition unit that acquires status information, which is information about the status of a machine tool to be controlled; a context information creation unit that creates, based on the status information, context information, which is information that characterizes the operating status of the machine tool and the operating status of the machine tool; a notification information creation unit that creates, based on the status information, notification information, which is information to be notified to a user of the machine tool when using the machine tool; a notification determination unit that determines a level of urgency of notification of the notification information based on the notification information and the context information; a display unit that determines a display method for the notification information based on the determined level of urgency, and displays the notification information using the determined display method; and an evaluation acquisition unit that acquires the user's evaluation of the displayed notification information, wherein the notification determination unit changes the level of urgency using the context information and the evaluation.

2. The control device according to claim 1, further comprising a cluster identification unit that clusters the notification information consisting of multiple pieces of notification information according to the similarity of the notification information and identifies the notification information cluster to which the notification information belongs, and the notification determination unit determines the urgency based on the notification information cluster identified by the cluster identification unit, and changes the urgency using the notification information cluster, the context information, and the evaluation.

3. The control device described in claim 1 or 2, characterized in that the context information includes a short context representing a feature of a first period among the features and a long context representing a feature of a second period among the features that is longer than the first period, and the notification determination unit determines the urgency based on the short context and the long context, and changes the urgency using the short context, the long context, and the evaluation.

4. A control device as described in any one of claims 1 to 3, further comprising a user identification unit that identifies user information that is information about the user, and the notification determination unit switches the urgency level based on the user information.

5. A control device as described in any one of claims 1 to 4, further comprising an evaluation inference unit that extracts state features, which are features extracted based on the state information, and infers an evaluation inference value, which is an inference value of the evaluation, based on the state features, and the evaluation acquisition unit acquires the evaluation using the evaluation inference value.

6. The control device according to claim 5, characterized in that the state feature quantity includes at least one of the following information: the time from when the notification information is displayed until the machine tool enters a specific state, the frequency of changes in the screen state before and after the notification information is displayed, and whether or not a specific operation is performed after the notification information is displayed.

7. A control device as described in any one of claims 1 to 6, characterized in that the notification information creation unit comprises: an abnormality judgment unit that judges whether an abnormality or warning has occurred in the machine tool based on changes in the status information; and a treatment information creation unit that creates treatment information indicating treatment content corresponding to the content based on the content of the abnormality or warning included in the judgment result of the abnormality judgment unit; and the display unit displays the notification information including the treatment information.

8. A control device as described in any one of claims 1 to 7, characterized in that the notification determination unit, when displaying the notification information next time, determines a display method for the notification information based on the changed urgency level, and displays the notification information using the determined display method, thereby displaying the notification information using a display method that reflects the changed urgency level.

9. A control method comprising: a status acquisition step in which a control device that controls a machine tool to be controlled acquires status information that is information about the status of the machine tool; a context information creation step in which the control device creates context information that is information that represents the operating status of the machine tool and the characteristics of the operating status of the machine tool based on the status information; a notification information creation step in which the control device creates notification information that is information to be notified to a user of the machine tool when using the machine tool based on the status information; a notification determination step in which the control device determines a level of urgency of the notification information based on the notification information and the context information; a display step in which the control device determines a display method for the notification information based on the determined level of urgency and displays the notification information using the determined display method; and an evaluation acquisition step in which the control device acquires an evaluation of the user's evaluation of the displayed notification information, wherein in the notification determination step, the control device changes the level of urgency using the context information and the evaluation.

Images