Maintenance support device and maintenance support method

Abstract

To provide maintenance support devices and other equipment that enable maintenance personnel to deepen their understanding of equipment maintenance. [Solution] The maintenance support device 10 includes a priority setting unit 145 that sets the priority for maintenance personnel to perform unchecked inspection items included in the equipment inspection checklist 132 based on a diagnostic knowledge database 11 in which data on fault diagnosis of the equipment to be maintained is accumulated, and a priority explanation generation unit 146 that generates an explanatory text including a recommended action that maintenance personnel should take after the most recent inspection, and the reason for prioritizing the recommended action over others, based on the diagnostic knowledge database 11 and the priority.

Description

Technical Field

[0001] The present disclosure relates to a maintenance support device and a maintenance support method.

Background Art

[0002] As a technology for supporting maintenance of equipment by maintenance personnel, for example, the technology described in Patent Document 1 is known. That is, Patent Document 1 describes a failure diagnosis device including an "inference control unit that narrows down candidates for interrogation based on the above priorities and preferentially executes interrogation regarding the user."

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technology described in Patent Document 1, candidates for interrogation are narrowed down based on priorities, but the basis for setting the priorities is not shown to the maintenance personnel. Then, for example, when a less experienced maintenance personnel performs equipment inspection, there is a possibility that the maintenance personnel cannot understand why the equipment is inspected in that priority (order of inspection).

[0005] Therefore, an object of the present disclosure is to provide a maintenance support device or the like that enables maintenance personnel to deepen their understanding of equipment maintenance.

Means for Solving the Problems

[0006] To solve the aforementioned problems, the maintenance support device according to this disclosure includes a priority setting unit that sets the priority for maintenance personnel to perform unchecked inspection items included in the inspection checklist of the equipment, based on a diagnostic knowledge database in which data on fault diagnosis of the equipment to be maintained is accumulated, and a priority explanation generation unit that generates an explanatory text including a recommended action that maintenance personnel should take after the most recent inspection, and the reason for prioritizing the recommended action over others, based on the diagnostic knowledge database and the priority. [Effects of the Invention]

[0007] According to this disclosure, it is possible to provide maintenance support devices, etc., that enable maintenance personnel to deepen their understanding of equipment maintenance. [Brief explanation of the drawing]

[0008] [Figure 1] This is an explanatory diagram including a maintenance support device according to an embodiment. [Figure 2] This is an example of the display of the top event selection screen in the maintenance support device according to the embodiment. [Figure 3] This is an example of a checklist display in a maintenance support device according to the embodiment. [Figure 4] This is a functional block diagram of the maintenance support device according to the embodiment. [Figure 5] This is a flowchart showing the processing performed by the processing unit of the maintenance support device according to the embodiment. [Figure 6] This is an explanatory diagram showing an example of a grid connection graph included in the diagnostic knowledge database of the maintenance support device according to the embodiment. [Figure 7] This is an explanatory diagram showing an example of an FMEA graph included in the diagnostic knowledge database of the maintenance support device according to the embodiment. [Figure 8] This is an explanatory diagram showing an example of a functional failure tree included in the diagnostic knowledge database of the maintenance support device according to the embodiment. [Figure 9] This is an explanatory diagram showing an example of a diagnostic tree included in the diagnostic knowledge database of the maintenance support device according to the embodiment. [Figure 10] This is an explanatory diagram showing an example of a diagnostic history database for a maintenance support device according to an embodiment. [Figure 11] This is an explanatory diagram of the data table used to set the checklist for the maintenance support device according to the embodiment. [Figure 12] This is an explanatory diagram of a data table showing the status of checks by maintenance personnel in a maintenance support device according to the embodiment. [Figure 13] This is an explanatory diagram showing how inspection results are linked to the functional failure tree of the maintenance support device according to the embodiment. [Figure 14] This is a flowchart showing the processing of the checklist update unit of the maintenance support device according to the embodiment. [Figure 15] This is an example of displaying explanatory text regarding the progress of inspections and the setting of priorities in a maintenance support device according to the embodiment. [Figure 16] This is another example of how explanatory text regarding the progress of inspections and the setting of priorities can be displayed in the maintenance support device according to the embodiment. [Figure 17] This figure shows the hardware configuration of the maintenance support device according to the embodiment. [Modes for carrying out the invention]

[0009] <<Embodiment>> Figure 1 is an explanatory diagram including a maintenance support device 10 according to an embodiment. The maintenance support device 10 shown in Figure 1 is a device for supporting the maintenance of equipment A1 by maintenance personnel M1. Here, "maintenance" of equipment A1 includes tasks such as inspections, identifying the cause of malfunctions, and taking measures such as maintenance. Examples of equipment A1 to be maintained include refrigeration equipment, air conditioning equipment, and industrial machinery, but are not limited to these. In addition, equipment A1 may include infrastructure such as power transmission equipment, substations, communication equipment, gas equipment, water supply equipment, railways, and roads, as well as plants such as power plants, manufacturing plants, water treatment plants, and chemical plants.

[0010] The inspection of such equipment A1 is performed, for example, when a malfunction occurs in equipment A1, but is not limited thereto, and may be a regular inspection. The maintenance support device 10 executes a predetermined process based on the input operation of the maintenance staff M1 and transmits the processing result to the display device 20. The display device 20 displays the processing result of the maintenance support device 10 and is configured to perform predetermined communication with the maintenance support device 10. As such a display device 20, for example, a liquid crystal display may be used, or a touch panel type portable terminal such as a smartphone or a tablet may be used. In the following description, the reference numerals of the equipment A1 and the maintenance staff M1 will be appropriately omitted.

[0011] FIG. 2 is a display example of the selection screen D1 for the top event. Hereinafter, as an example, a case where a customer requests repair because the outside air cannot be sent by the blower will be described. In such a case, the maintenance staff goes to the site where the refrigerator to be maintained is installed, and first selects the top event via the touch panel type display device 20 (see FIG. 1). Here, the "top event" is a major classification indicating the content of the malfunction of the equipment and is preset. In the example of FIG. 2, as the top events related to the malfunction of the refrigerator, items such as "cooling trouble", "blower trouble", and "compressor trouble" are displayed vertically on the display device 20 (see FIG. 1).

[0012] The maintenance staff selects the top event based on the situation description from the customer and the condition of the site. In the example of FIG. 2, "blower trouble" is selected as the top event. The selection result of the top event by the maintenance staff is transmitted to the maintenance support device 10 (see FIG. 1) via a network or the like. The maintenance support device 10 sets a checklist 132 (see FIG. 3) corresponding to the content of the top event selected by the maintenance staff. Here, the "checklist" is a list in which a plurality of inspection items when the maintenance staff inspects the equipment are listed.

[0013] FIG. 3 is a display example of the checklist 132. For example, maintenance personnel inspect the refrigeration unit (equipment) while referring to a checklist 132 as shown in Figure 3, and reflect the results of the inspection in the checklist 132 as appropriate. Also, "inspection" means that maintenance personnel check each designated part of the equipment to see if there are any malfunctions. As mentioned above, the checklist 132 is pre-set and corresponds to a predetermined top event. In the example in Figure 3, three inspection items are displayed vertically on the display device 20 (see Figure 1) corresponding to the top event of "blower trouble" in the refrigeration unit (see Figure 2).

[0014] As will be explained in more detail later, in this embodiment, inspection items are displayed in order of priority, from highest to lowest. In the example in Figure 3, the inspection item "No. 1," "Check if it rotates manually," has the highest priority and is therefore displayed at the top of the screen. In addition, the inspection item "No. 2," which has the second highest priority, is displayed second from the top of the screen, and the inspection item "No. 3," which has the third highest priority, is displayed third from the top.

[0015] The "Judgment" shown in Figure 3 indicates the inspection results of the maintenance worker, and is entered for each inspection item. Specifically, the maintenance worker selects whether the inspected area is normal or abnormal through their input. By using such a checklist 132, inspection omissions and errors by the maintenance worker can be suppressed. In addition, each time the maintenance worker performs a check, the display order of the inspection items (i.e., the order of priority) is updated appropriately by the maintenance support device 10 (see Figure 1).

[0016] Figure 4 is a functional block diagram of the maintenance support device 10. As shown in Figure 4, the maintenance support device 10 includes a diagnostic knowledge database 11, a diagnostic history database 12, a UI unit 13, and a processing unit 14. The diagnostic knowledge database 11 is a database that stores data related to fault diagnosis of the equipment to be maintained. "Fault diagnosis" means identifying what kind of fault has occurred in the equipment and what its cause is. The diagnostic history database 12 is a database that stores the history of fault diagnosis of the equipment.

[0017] The UI unit 13 is an interface for data input by maintenance personnel and data output to the display device 20 (see Figure 1). "UI" is an abbreviation for "User Interface." As shown in Figure 4, the UI unit 13 includes an event input unit 131, a checklist 132, an explanation display unit 133, and a diagnostic result input unit 134.

[0018] The event input unit 131 is an input interface used by maintenance personnel to select the top event (for example, "blower trouble") as described in Figure 2. The checklist 132 (see also Figure 3) is a list containing multiple inspection items, as described above, and is displayed on the display device 20 (see Figure 1) in a predetermined manner.

[0019] The explanatory display unit 133 is an output interface for displaying the progress of inspections by maintenance personnel, as well as recommended actions that maintenance personnel should take next and the rationale for those actions, as predetermined explanatory text. The diagnostic result input unit 134 is an input interface for maintenance personnel to input the results of fault diagnosis.

[0020] The processing unit 14 executes predetermined processes based on predetermined programs or input operations by maintenance personnel. As shown in Figure 4, the processing unit 14 includes a checklist setting unit 141, a checklist update unit 142, an inspection progress status identification unit 143, an inspection progress status explanation generation unit 144, a priority setting unit 145, and a priority explanation generation unit 146.

[0021] The checklist setting unit 141 sets a checklist 132 (see also Figure 3) corresponding to the top event selected via the event input unit 131 (for example, "Blower trouble" in Figure 2). The checklist update unit 142 updates the order in which unchecked inspection items are displayed on the display device 20 (see Figure 1) based on a predetermined priority whenever a maintenance worker checks the items on the checklist 132.

[0022] The inspection progress identification unit 143 identifies the inspection progress based on the check status of the checklist 132. The inspection progress explanation generation unit 144 generates an explanatory text regarding the progress of the equipment inspection by the maintenance personnel based on the check status of the checklist 132 and the diagnostic knowledge database 11.

[0023] The priority setting unit 145 sets the priority for maintenance personnel to perform unchecked inspection items included in the equipment inspection checklist 132, based on the diagnostic knowledge database 11. The priority explanation generation unit 146 generates an explanation text based on the diagnostic knowledge database 11 and the priority, which includes the recommended action that maintenance personnel should take after the most recent inspection, and the reasoning for prioritizing that recommended action over others.

[0024] Figure 5 is a flowchart illustrating the processes performed by the processing unit of the maintenance support device (see also Figure 4 as appropriate). Furthermore, when "START" is displayed in Figure 5, it is assumed that a predetermined top event related to equipment malfunction (for example, "Blower trouble" in Figure 2) has already been selected by the maintenance personnel. In step S101, the processing unit 14 sets the checklist 132 (see Figure 3) using the checklist setting unit 141. As described above, the checklist 132 is set in accordance with the predetermined top event selected by the maintenance worker.

[0025] In step S102, the processing unit 14 determines whether a new check has been entered in the checklist 132. If no new check has been entered in the checklist 132 (S102: No), the processing unit 14 repeats the process in step S102. If a new check has been entered in the checklist 132 (S102: Yes), the processing unit 14 proceeds to step S103. In the example in Figure 3, the maintenance worker has selected "abnormal" as the result for the inspection item "check if the blower fan rotates manually" (i.e., a new check has been entered).

[0026] In step S103, the processing unit 14 uses the inspection progress identification unit 143 to identify the progress of the inspection by the maintenance personnel. Specifically, the processing unit 14 identifies how far the equipment inspection has progressed based on the check status of the checklist 132 (see Figure 3). In step S104, the processing unit 14 sets the priority of the unchecked inspection items using the priority setting unit 145 (priority setting process). Details of the priority setting will be described later.

[0027] In step S105, the processing unit 14 updates the checklist 132 using the checklist update unit 142. That is, the processing unit 14 ensures that the unchecked inspection items are displayed as the checklist 132 in the order of priority set in step S104.

[0028] In step S106, the processing unit 14 generates an explanatory text about the progress of the inspection using the inspection progress explanation generation unit 144. In step S107, the processing unit 14 generates an explanatory text regarding the inspection priority using the priority explanation generation unit 146 (priority explanation generation process). Details of steps S106 and S107 will be described later.

[0029] In step S108, the processing unit 14 displays the checklist 132, as well as explanatory text regarding the progress and priority of the inspection, on the display device 20 via the UI unit 13. In step S109, the processing unit 14 determines whether the maintenance by the maintenance personnel has been completed. The completion of maintenance is determined, for example, based on whether the end button (not shown) on the display screen has been selected. If the maintenance is not completed in step S109 (S109: No), the processing unit 14 returns to step S102.

[0030] Each time a maintenance worker checks an item on the checklist 132, the result of that check (the data in the "Judgment" column shown in Figure 3) is transmitted to the maintenance support device 10 via the network or the like. Although not shown in Figure 5, the processing unit 14 of the maintenance support device 10 narrows down the equipment failure mode (type of failure) based on the diagnostic knowledge database 11, identifies the cause of each failure mode, and then displays the appropriate countermeasures on the display device 20. If the maintenance worker takes the prescribed action based on these countermeasures and the equipment maintenance is completed (S109: Yes), the processing unit 14 proceeds to step S110. In step S110, the processing unit 14 reflects the diagnostic results regarding the equipment in the diagnostic history database 12 and terminates the series of processes (END).

[0031] Note that the flowchart shown in Figure 5 is just one example and is not limited to this. For example, the processing unit 14 may generate an explanatory text about the inspection progress immediately after identifying the inspection progress (S103) (S106). Alternatively, the processing unit 14 may update the checklist 132 after generating explanatory texts about the inspection progress and priority (S106, S107) (S105).

[0032] The diagnostic knowledge database 11 (see Figure 4) pre-stores data related to equipment fault diagnosis as a predetermined graph database. Here, a "graph database" is a database in which a graph containing nodes (points) and edges (lines) is represented in a predetermined data structure. By using such a graph database, it becomes possible to efficiently represent and search for relationships between data. The diagnostic knowledge database 11 pre-stores data such as that shown in Figures 6 to 9 below.

[0033] Figure 6 is an explanatory diagram showing an example of a system connection graph G1 included in the diagnostic knowledge database. Note that the system connection graph G1 shown in Figure 6 is a graph showing the connection relationships of the main components (equipment) of the equipment (refrigeration unit in the example of Figure 6), and is pre-configured. Nodes N1 to N4 shown in Figure 6 are nodes representing the main components of the refrigeration unit, corresponding in this order to the compressor, condenser, expansion valve, and evaporator.

[0034] Edge E1 indicates that high-pressure gaseous refrigerant is led from the compressor to the condenser via refrigerant piping. Edge E2 indicates that high-pressure liquid refrigerant is led from the condenser to the expansion valve via refrigerant piping. Edge E3 indicates that low-pressure liquid refrigerant is led from the expansion valve to the evaporator via refrigerant piping. Edge E4 indicates that low-pressure gaseous refrigerant is led from the evaporator to the compressor via refrigerant piping.

[0035] In the example shown in Figure 6, the refrigerant system is represented by the system connection graph G1, but this is not the only way to represent it. For example, electrical systems and power transmission systems may also be shown in a given system connection graph.

[0036] Figure 7 is an explanatory diagram showing an example of FMEA graph G2 included in the diagnostic knowledge database. The FMEA graph G2 shown in Figure 7 is a pre-configured graph that associates data such as the function, failure mode, and inspection items of the equipment components (devices). "FMEA" is an abbreviation for "Failure Mode and Effects Analysis." Nodes N11 and N12 in Figure 7 represent the chiller and its main component, the evaporator. Edge E11, which connects nodes N11 and N12, indicates the inclusion relationship between nodes N11 and N12 (chiller and evaporator). In the example in Figure 7, the identifier "hasSubComponent" is associated with edge E11, indicating that the evaporator is a component of the chiller.

[0037] Node N13 indicates the function of node N12 (evaporator), which is connected via edge E12. Specifically, it is shown that the evaporator indicated by node N12 has the function of evaporating the refrigerant and cooling the air. Node N14 indicates the failure mode of node N12 (evaporator), which is connected via edge E13. Specifically, it is shown that the evaporator indicated by node N12 has a failure mode of filter fouling.

[0038] Node N15 indicates the cause of node N14 (filter fouling), which is connected via edge E14. Specifically, it indicates that filter fouling occurs due to insufficient cleaning of the filter. Node N16 indicates the nature of the functional failure of node N13 (evaporator function), which is connected via edge E15. Specifically, it indicates that a functional failure occurs in which no cold air is generated when the function of evaporating the refrigerant and cooling the air is lost. Edge E16, which connects nodes N14 and N16, indicates that a functional failure occurs in which no cold air is generated as a result of filter fouling.

[0039] Node N17 shows the effect of node N16 (faulty), which is connected via edge E17. Specifically, it shows that the failure to generate cold air has the effect of increasing the internal temperature. Node N18 shows the check points related to node N16 (faulty), which is connected via edge E18. Specifically, it shows that the refrigerant flow should be checked to see if a failure to generate cold air is occurring.

[0040] Figure 8 is an explanatory diagram showing an example of a functional failure tree G3 included in the diagnostic knowledge database of a maintenance support device. The functional fault tree G3 shown in Figure 8 is a graph that shows the relationship between events associated with equipment failure and inspection items to check for the presence or absence of those events, and is pre-configured. In the example in Figure 8, the functional fault tree G3 is represented as a tree structure graph in the form of a fault tree. Node N21 shown in Figure 8 indicates the top event. As mentioned above, the top event is a broad category indicating the content of the equipment malfunction, and is selected by the maintenance personnel from among several candidates (see Figure 2). For example, if "blower trouble" is selected as the top event, the functional fault tree G3 with "blower trouble" as the top event node N21 is read from the diagnostic knowledge database 11 (see Figure 4).

[0041] Nodes N22, N23, and N24 shown in Figure 8 represent intermediate events (functional failures) that may occur as a result of the aforementioned top event (blower trouble). These nodes N22, N23, and N24 are located below node N21, the top event node, in the tree-structured functional failure tree G3.

[0042] Node N31 is a logic gate that performs an OR operation. In the example in Figure 8, it is shown that if the blower fan cannot rotate (node ​​N23) or if air circulation cannot be performed (node ​​N24), an intermediate event (node ​​N22) occurs in which outside air cannot be supplied.

[0043] Nodes N32, N33, and N34 indicate inspection items and are connected via edges to predetermined nodes for intermediate events. For example, to check whether the intermediate event (node ​​N22) of being unable to supply outside air has occurred, an inspection item (node ​​N32) is set to check whether the blower fan rotates manually with the power off. Nodes N35 and N36 indicate the failure modes that cause the intermediate events. For example, if the intermediate event (node ​​N23) of the blower fan being unable to rotate occurs, the failure mode indicated as the cause is bearing failure.

[0044] Figure 9 is an explanatory diagram showing an example of diagnostic tree G4 included in the diagnostic knowledge database. The diagnostic tree G4 shown in Figure 9 is a tree-structured graph that shows the relationship between equipment inspection items and failure modes. Such a diagnostic tree G4 may, but is not limited to, be pre-stored in the diagnostic knowledge database 11 (see Figure 4). For example, the processing unit 14 may generate the diagnostic tree G4 based on the functional failure tree G3 (see Figure 8). The diagnostic tree G4 is configured such that the intermediate event nodes N22, N23, and N24 of the functional failure tree G3 (see Figure 8) are replaced in that order with the inspection item nodes N32, N33, and N34.

[0045] In the example in Figure 9, for the top event, a blower malfunction (node ​​N21), a maintenance worker checks whether the blower fan rotates manually (node ​​N32). If an abnormality is found, the maintenance worker performs a manual inspection of the motor bearings (node ​​N33) and a visual inspection of the fan (node ​​N34). Such a diagnostic tree G4 is used to generate checklist 132 (see Figure 3).

[0046] Figure 10 is an explanatory diagram showing an example of a diagnostic history database 12. As mentioned above, the diagnostic history database 12 shown in Figure 10 is a database that shows the history of equipment failure diagnoses. In the example in Figure 10, the diagnostic history database 12 associates the start date and time of the maintenance work with the end date and time of the maintenance work, the name of the customer who requested the maintenance work, the customer's location, the equipment (equipment component) that was the subject of the maintenance work, its symptoms, the inspection process, and the actions taken by the maintenance personnel.

[0047] The "Symptoms" shown in Figure 10 refer to the specific details of the symptoms related to the equipment failure (the intermediate events mentioned above). For example, the information provided by the customer when requesting maintenance work is entered in this "Symptoms" column. The "Process" is linked to the results of the maintenance worker's checklist 132 (see Figure 3). The "Action Taken" is the actual action taken by the maintenance worker on the equipment.

[0048] In the example in Figure 10, when a device called Asset4 was unable to receive outside air, a maintenance worker performed a prescribed Process A inspection (e.g., visual inspection of the fan) and found that the blower fan was broken. The solution was to replace the blower fan. Similarly, when a device called Asset1 was unable to receive power, a maintenance worker performed a prescribed Process B inspection and the solution was to replace the power supply.

[0049] Next, we will explain, in order, how to set up Checklist 132 (see Figure 3), how to identify the progress of the inspection, how to update Checklist 132, and how to generate explanatory text.

[0050] Figure 11 is an explanatory diagram of data table T1 used for setting up the checklist. The data table T1 shown in Figure 11 is generated when a predetermined top event (for example, "blower trouble": see Figure 2) is input via the event input unit 131 (see Figure 4). Such a data table T1 is generated, for example, by the checklist setting unit 141 (see Figure 4) based on the diagnostic tree G4 (see Figure 9) described above.

[0051] In data table T1, "No.", "Node ID", and "Inspection Item" are associated. "No." is the number assigned to the inspection item, but it also indicates the priority order when maintenance personnel perform the inspection item. "Node ID" is the identification information of the node corresponding to the inspection item, and it is associated with a specific node in each graph (see Figures 6 to 9) of the diagnostic knowledge database 11 (see Figure 4).

[0052] For example, node ID "n1" shown in Figure 11 corresponds to node N32 in diagnostic tree G4 (see Figure 9). Similarly, node ID "n2" corresponds to node N33 in diagnostic tree G4, and another node ID "n3" corresponds to node N34 in diagnostic tree G4.

[0053] The checklist setting unit 141 (see Figure 4) generates a data table T1 based on the diagnostic tree G4 (see Figure 9), and sets the checklist 132 (see Figure 3) based on this data table T1. The checklist 132 set in this manner is updated to a predetermined value by the checklist update unit 142 (see Figure 4) each time a maintenance worker checks an inspection item.

[0054] Figure 12 is an explanatory diagram of data table T2, which shows the status of checks by maintenance personnel. The data table T2 shown in Figure 12 indicates the status of the maintenance worker's checks on checklist 132 (see Figure 3) (i.e., how far the inspection has progressed), and is generated by the processing unit 14 (see Figure 4). In the example in Figure 12, "No.", "Node ID", "Inspection Item", "Normal", and "Abnormal" are associated in data table T2. Note that "No.", "Node ID", and "Inspection Item" included in data table T2 are the same as those in data table T1 (see Figure 11) described above, so their explanation is omitted.

[0055] The "Normal" and "Abnormal" columns in data table T2 reflect the results of the normal or abnormal determination in checklist 132 (see Figure 3) (the result of the input operation by the maintenance worker). Specifically, if "Normal" is selected in checklist 132, the "Normal" column will be "True" and the "Abnormal" column will be "False". Conversely, if "Abnormal" is selected in checklist 132, the "Normal" column will be "False" and the "Abnormal" column will be "True".

[0056] In the example in Figure 12, the maintenance worker inspected whether the blower fan rotated manually, and as a result, "abnormal" was selected in checklist 132 (see Figure 3). Therefore, the "Normal" column is set to "False" and the "Abnormal" column is set to "True". Note that the manual inspection of the motor bearings and the visual inspection of the fan were not checked in checklist 132 (see Figure 3), so both the "Normal" and "Abnormal" columns are set to "None".

[0057] The inspection progress identification unit 143 (see Figure 4) links the node IDs of each inspection item in data table T2 to the FMEA graph G2 (see Figure 7), the functional failure tree G3 (see Figure 8), and the diagnostic tree G4 (see Figure 9), as well as the inspection results of the maintenance personnel.

[0058] Figure 13 is an explanatory diagram showing how inspection results are linked to the functional failure tree G3R. Note that each node and edge of the functional failure tree G3R is the same as in the functional failure tree G3 in Figure 8. The inspection progress identification unit 143 (see Figure 4) associates the maintenance worker's inspection results with predetermined nodes based on the data table T2 (see Figure 12). For example, in the data table T2 (see Figure 12), the node ID for the inspection item "Check if it rotates manually" is n1 (corresponding to node N32 in Figure 13), and the inspection result is "True" in the "Abnormal" column. In such a case, the inspection progress identification unit 143 associates node N32R, which is "Abnormal" as an inspection result, with node N32 of the inspection item "Check if it rotates manually".

[0059] This allows the processing unit 14 to understand that, in response to the intermediate event of being unable to supply outside air (node ​​N22), a maintenance worker checks whether the blower fan rotates manually (node ​​N32), and that the result was abnormal (Abnormal: node N32R). Note that the inspection item "manual inspection of motor bearings" has not been inspected (see Figure 12), so node N33 for this inspection item is associated with node N33R, which is "None" (the same applies to other nodes N34R).

[0060] In the example in Figure 13, the progress of the inspection is represented using a functional failure tree G3R, but this is not the only option. For example, instead of (or in conjunction with) the functional failure tree G3R, it is also possible to represent the progress of the inspection using an FMEA graph G2 (see Figure 7) or a diagnostic tree G4 (see Figure 9).

[0061] As described above, the priority setting unit 145 (see Figure 4) sets the priority of the unchecked inspection items included in the checklist 132 (see Figure 3). Specifically, the priority setting unit 145 provides the generating AI (Artificial Intelligence) with context from data such as the functional failure tree G3R (see Figure 13) to which the inspection results are linked, as well as the system connection graph G1 (see Figure 6), the FMEA graph G2 (see Figure 7), the diagnostic tree G4 (see Figure 9), and the diagnostic history database 12 (see Figure 10). Examples of such generating AI include GraphRAG and Think-on-Graph, which are capable of reading graph databases. The priority setting unit 145 then causes the generating AI to determine the priority of the unchecked inspection items whose inspection result is "None" (see Figure 13).

[0062] Furthermore, the following are examples of criteria used by the priority setting unit 145 (see Figure 4) when setting the priority of the inspection order. For example, the priority setting unit 145 may set a higher priority for inspection items that require less man-hours to inspect among the uninspected inspection items. In this case, the diagnostic knowledge database 11 (see Figure 4) is assumed to store data on the man-hours required for each of the multiple inspection items related to the equipment. As a result, inspection items that require less man-hours are carried out preferentially, thereby improving the efficiency of maintenance work. Incidentally, it is also possible for the priority setting unit 145 to read the average amount of man-hours actually required in past inspections from the diagnostic history database 12 (see Figure 4) and set the inspection priority based on the average amount of man-hours.

[0063] Furthermore, the priority setting unit 145 (see Figure 4) may be configured to prioritize inspection items that have a higher failure frequency or probability of failure at the inspection location among the uninspected inspection items. In this case, the diagnostic knowledge database 11 (see Figure 4) stores data on the failure frequency or probability of failure at the inspection location for each of the multiple inspection items related to the equipment. This allows for, for example, prioritizing the execution of inspection items with a high probability of failure, thereby improving the efficiency of maintenance operations.

[0064] Furthermore, the priority setting unit 145 (see Figure 4) may be configured to prioritize inspection items that are less difficult to inspect among the uninspected items. In this case, the diagnostic knowledge database 11 (see Figure 4) stores data indicating the difficulty level of each of the multiple inspection items related to the equipment. This ensures that inspections with lower difficulty levels are prioritized, thus preventing excessive time from being wasted on inspections.

[0065] Furthermore, the priority setting unit 145 (see Figure 4) may, if the inspection points of the inspection items that have been most recently checked by a maintenance worker in the checklist 132 (see Figure 3) are included in a predetermined system, set the priority of other inspection points included in that predetermined system higher than the inspection points in systems other than that predetermined system. In this case, the diagnostic knowledge database 11 (see Figure 4) is assumed to store identification information of the system containing the inspection points for each of the multiple inspection items related to the equipment. By setting priorities based on the grouping of systems in this way, a maintenance worker can inspect multiple inspection points in a single system at once. Examples of such systems include water systems, electrical systems, and refrigerant systems.

[0066] Furthermore, the priority setting unit 145 (see Figure 4) may, based on the diagnostic history database 12 (see Figure 4) in which the equipment's diagnostic history is stored, set a lower priority for uninspected inspection items for which a predetermined period has not elapsed since the last inspection, compared to other inspection items for which a predetermined period has elapsed. This is because the likelihood of finding an abnormality in an inspection location for which a predetermined period has not elapsed since the last inspection is low.

[0067] Furthermore, the priority setting unit 145 (see Figure 4) may, based on the diagnostic history database 12 (see Figure 4) in which the equipment's diagnostic history is accumulated, set the priority of the uninspected inspection items that were inspected earlier in the past by a skilled maintenance worker to be higher than other inspection items. By referring to the inspection history of skilled maintenance workers in this way, the efficiency of maintenance work can be improved. It should be assumed that data indicating the skill level of maintenance workers, associated with the identification information of maintenance workers, is pre-stored in the diagnostic history database 12.

[0068] Furthermore, the aforementioned multiple criteria (criteria for setting priorities) may be combined as appropriate. The priorities set by the priority setting unit 145 are used as appropriate to update the checklist 132 (see Figure 3), as will be explained below.

[0069] Figure 14 is a flowchart showing the processing of the checklist update section (see also Figure 4 as appropriate). Figure 14 shows the details of the process in step S105 of Figure 5, which is performed, for example, when a maintenance worker is conducting an inspection. Also, it is assumed that checklist 132 (see Figure 3) has already been set when "START" is displayed in Figure 14.

[0070] In step S1051, the checklist update unit 142 retrieves inspection items from data table T2 (see Figure 12). In the example in Figure 12, three inspection items are retrieved from data table T2: "check if it rotates manually", "manual inspection of motor bearings", and "visual inspection of fan".

[0071] In step S1052, the checklist update unit 142 obtains the check status of checklist 132 (see Figure 3). Specifically, the checklist update unit 142 refers to the "Normal" and "Abnormal" columns of data table T2 (see Figure 12) to determine the stage to which the inspection by the maintenance personnel has progressed.

[0072] In step S1053, the checklist update unit 142 acquires the priority. As described above, the priority for maintenance personnel to inspect the equipment is set by the priority setting unit 145.

[0073] In step S1054, the checklist update unit 142 rearranges the checklist 132 (see Figure 3) based on priority. For example, in the example in Figure 3, if the maintenance worker's inspection of whether the blower fan rotates manually is found to be abnormal, it may be better to visually check the fan before manually inspecting the motor bearings. When the order of priority changes in this way, the checklist update unit 142 rearranges the display order of the inspection items to a predetermined order.

[0074] Furthermore, items that no longer require inspection or have already been inspected may be removed from checklist 132. This makes it easier for maintenance personnel to identify uninspected items. The series of processes shown in Figure 14 are repeated as prescribed each time a maintenance worker checks "normal" or "abnormal" on checklist 132 (see Figure 3).

[0075] The inspection progress explanation generation unit 144 (see Figure 4) generates an explanatory text regarding the progress of the inspection. For example, the inspection progress explanation generation unit 144 provides the generating AI with a functional failure tree G3R (see Figure 13) to which the inspection results are linked as context. As such a generating AI, GraphRAG or Think-on-Graph as described above can be used as appropriate. Alternatively, the generating AI may be provided with an FMEA graph G2 (see Figure 7) or a diagnostic tree G4 (see Figure 9) to which the inspection results are linked. The explanatory text generated by the inspection progress explanation generation unit 144 is displayed on the display device 20 (see Figure 1) via the explanation display unit 133 (see Figure 4).

[0076] The priority explanation generation unit 146 (see Figure 4) generates an explanatory text regarding the setting of priorities. Specifically, the priority explanation generation unit 146 provides the generation AI with a functional failure tree G3R (see Figure 13), an FMEA graph G2 (see Figure 7), and a diagnostic tree G4 (see Figure 9) linked to the inspection results of inspection items as context. The priority explanation generation unit 146 then uses the generation AI to generate an explanatory text that includes the recommended action (inspection content, etc.) that the maintenance worker should take after the most recent inspection, and the rationale for prioritizing that recommended action over others. The explanatory text thus generated is displayed on the display device 20 (see Figure 1) via the explanatory display unit 133 (see Figure 4).

[0077] Figure 15 shows an example of explanatory text regarding the progress and prioritization of inspections. In the example in Figure 15, explanatory texts K1 and K2 from the check status in Figure 12 are displayed regarding the blower trouble of the refrigeration unit. The upper explanatory text K1 is an explanation of the inspection progress and is generated by the inspection progress explanation generation unit 144 (see Figure 4). As mentioned above, in data table T2 (see Figure 12), the result of the inspection of whether the blower fan rotates manually is abnormal, while the manual inspection of the motor bearing and the visual inspection of the fan are in an uninspected state (None).

[0078] Based on this data, the inspection progress explanation generation unit 144 explains the inspected items and the uninspected items separately in the explanation text K1. The explanation text, "In this case, bearing damage and blower fan fracture are possible," is generated appropriately based, for example, on the functional failure tree G3R in Figure 13. In other words, the aforementioned explanation text is generated based on the failure modes (nodes N35, N36) connected to the uninspected inspection items (nodes N33, N34) via edges, etc. By reading such explanation texts, maintenance personnel can understand what has been inspected so far and what failure modes are possible.

[0079] The explanatory text K2 shown in Figure 15 is an explanation of the maintenance worker's recommended actions and priorities, and is generated by the priority explanation generation unit 146 (see Figure 4). Specifically, the explanatory text displayed is, "Since the fan breakage can be visually confirmed, we recommend that you first visually check the fan." The latter part of this explanatory text (visually check the fan) indicates the maintenance worker's recommended action, indicating that visually checking the fan has the highest priority. Furthermore, the preceding part of the explanatory text (the fan breakage can be visually confirmed) shows the basis for giving the highest priority to visually checking the fan. By reading such explanatory text, maintenance workers can understand the reason why they should visually check the fan first, thereby deepening their understanding of maintenance work.

[0080] Figure 16 shows another example of how explanatory text regarding the progress and prioritization of inspections can be displayed. In the example in Figure 16, explanatory texts K3 and K4 are displayed for the case where the refrigerator's "cooling function is insufficient." Explanatory text K3 is an explanation of the inspection progress and is generated by the inspection progress explanation generation unit 144 (see Figure 4). Specifically, it explains the inspected items (power system inspection) and the uninspected items separately.

[0081] Explanatory text K4 is an explanation of recommended actions and priorities for maintenance personnel, and is generated by the priority explanation generation unit 146 (see Figure 4). The first sentence of explanatory text K4 explains the function of the evaporator (corresponding to node N13 in Figure 7), and the second sentence shows the failure modes that cause functional failure of the evaporator (corresponding to node N14, etc., in Figure 7). These explanatory texts are generated, for example, based on the FMEA graph G2 (see Figure 7).

[0082] The third sentence of explanatory text K4 explains that, regarding the unchecked inspection item, the filter is unlikely to be dirty because it was recently replaced, and therefore a refrigerant shortage is more likely. The fact that the filter was replaced last month is identified based on the diagnostic history database 12 (see Figure 4).

[0083] The fourth sentence of explanatory text K4 explains that the evaporator receives refrigerant from the expansion valve via the low-pressure Freon piping. Such explanatory texts are generated based on the system connection graph G1 (see Figure 6). Furthermore, the fifth sentence of explanatory text K4 explains that the first thing to check is whether the refrigerant is being supplied correctly from the low-pressure Freon piping.

[0084] In this way, the priority explanation generation unit 146 (see Figure 4) generates explanatory text that shows the structural causal relationships related to the functional failure of the equipment, based on the diagnostic knowledge database 11. Although not specifically shown in Figure 16, it is also possible to display explanatory text that shows the aforementioned structural causal relationships, such as, "If there is a blockage in the low-pressure refrigerant piping, the evaporator connected downstream of the low-pressure refrigerant piping will lack refrigerant, making it difficult to generate cold air in the evaporator." Such explanatory text is generated based on the system connection graph G1 (see Figure 6) and the FMEA graph G2 (see Figure 7).

[0085] In this way, the recommended actions of maintenance personnel and the rationale behind them are explained, allowing even inexperienced maintenance personnel to efficiently carry out maintenance work by following the explanations. Furthermore, by reading the explanations, maintenance personnel can deepen their understanding of maintenance work. Note that the explanations shown in Figures 15 and 16 may be displayed on the display device 20 (see Figure 1) on a single screen together with the checklist 132 (see Figure 3), or they may be displayed independently; either is acceptable.

[0086] Figure 17 shows the hardware configuration of the maintenance support device 10. The maintenance support device 10 has a hardware configuration comprising a processor 10a, RAM 10b (Random Access Memory), ROM 10c (Read Only Memory), HDD 10d (Hard Disk Drive), a communication interface 10e, and an input / output interface 10f, all of which are predeterminedly connected via an internal bus 10g.

[0087] The processor 10a is hardware that constitutes the processing unit 14 (see Figure 4) of the maintenance support device 10. The RAM 10b, ROM 10c, and HDD 10d are hardware that constitute the storage unit (not shown) of the maintenance support device 10. The processor 10a reads a predetermined program stored in the ROM 10c or HDD 10d and loads it into the RAM 10b, thereby executing a predetermined process.

[0088] The communication interface 10e communicates with the display device 20 as specified via a network or the like (not shown). The input / output interface 10f performs data input and data output as specified when an input device such as a mouse or keyboard or a display device such as a display is connected to the maintenance support device 10.

[0089] Such a maintenance support device 10 may consist of a single computer, or it may consist of multiple computers connected via signal lines or a network. For example, the functions of the maintenance support device 10 may be distributed across multiple computers such as cloud servers or edge servers.

[0090] According to this embodiment, the recommended action that the maintenance worker should take next and the reasoning behind it are displayed on the display device 20, so even inexperienced maintenance workers can efficiently carry out equipment maintenance work. In addition, since maintenance workers can perform inspections while understanding the reasoning behind why certain inspection items should be carried out before others, their understanding of maintenance work can be deepened.

[0091] ≪Variations≫ Although embodiments of the maintenance support device 10 and maintenance support method related to this disclosure have been described above, this disclosure is not limited to these descriptions, and various modifications can be made. For example, in the embodiment described, a case was described in which the grid connection graph G1 (see Figure 6) and the FMEA graph G2 (see Figure 7) are provided separately, but the embodiment is not limited to this. That is, the grid connection graph and the FMEA graph may be coupled in a predetermined manner.

[0092] Furthermore, although the embodiment described a case in which multiple candidate top events are displayed on the display device 20 (see Figure 2), it is not limited to this. For example, a maintenance worker may input a predetermined event (such as the status of equipment) in text or voice, and the generating AI may select the event that is closest to the event input by the maintenance worker from among the multiple top events.

[0093] Furthermore, while the embodiment described the case in which the system connection graph G1 (see Figure 6), FMEA graph G2 (see Figure 7), functional failure tree G3 (see Figure 8), and diagnostic tree G4 (see Figure 9) are used in the internal processing of the maintenance support device 10, it is not limited to this. That is, each of the above graphs may be displayed on the display device 20 in a predetermined manner based on input operations by a maintenance worker.

[0094] Furthermore, while the embodiment described a case where the display order of inspection items is changed based on priority, it is not limited to this. For example, the number assigned to the inspection items may be changed based on priority, while the display order of the inspection items is maintained. Even with such processing, maintenance personnel can understand the priority of the inspection items.

[0095] Furthermore, while the embodiments described cases in which a generation AI is used for setting priorities and generating explanatory texts, the invention is not limited to this. In other words, priorities may be set and explanatory texts may be generated based on predetermined graph search rules or predetermined formats without specifically using AI.

[0096] Furthermore, the processing (maintenance support method) performed by the maintenance support device 10 may be executed as a predetermined program on a computer. The aforementioned program can be provided via a communication line, or it can be written to a predetermined recording medium and distributed.

[0097] Furthermore, this disclosure is not limited to the embodiments and includes various modifications. For example, the embodiments are described in detail for the purpose of clearly illustrating this disclosure and are not necessarily limited to having all the configurations described. Also, some of the configurations of the embodiments can be added, deleted, or replaced with other configurations.

[0098] Furthermore, each of the aforementioned configurations, functions, processing units, processing means, etc., may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. Alternatively, each of the aforementioned configurations, functions, etc., may be implemented in software by having the processor interpret and execute programs that realize each function. Information such as programs, tables, and files that realize each function can be stored in memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0099] Furthermore, the control lines and information lines shown are those deemed necessary for explanatory purposes, and not all control lines and information lines are necessarily shown in the actual product. In reality, it can be assumed that almost all components are interconnected. [Explanation of Symbols]

[0100] 10 Maintenance support equipment 11 Diagnostic Knowledge Database 12. Diagnostic History Database 13 UI section 14 Processing Unit 20 Display device 131 Event Input Section 132 Checklist 133 Explanation Display Section 134 Diagnostic Result Input Section 141 Checklist Setting Section 142 Checklist Update Section 143 Inspection Progress Status (Specific Department) 144 Inspection Progress Explanation Generation Unit 145 Priority setting section 146 Priority Explanation Generation Unit A1 Equipment G1 System Connection Graph G2 FMEA graph G3 Functional Failure Tree G4 Diagnostic Tree M1 maintenance worker S104 Step (Priority Setting Process) S107 Step (Priority Explanation Generation Process)

Claims

1. The system includes a priority setting unit that sets the priority for maintenance personnel to perform unchecked inspection items included in the inspection checklist of the equipment, based on a diagnostic knowledge database in which data on fault diagnosis of the equipment to be maintained is accumulated. A maintenance support device comprising a priority explanation generation unit that generates an explanatory text including a recommended action to be taken by a maintenance worker after the most recent inspection, and the rationale for prioritizing that recommended action over others, based on the diagnostic knowledge database and the priority.

2. Each time a maintenance worker checks an item on the checklist, the checklist update unit updates the order in which unchecked items are displayed on the display device based on the priority order. A maintenance support device according to claim 1, characterized by the following:

3. The priority explanation generation unit generates an explanatory text showing the structural causal relationship regarding the functional failure of the equipment, based on the diagnostic knowledge database. A maintenance support device according to claim 1, characterized by the following:

4. The diagnostic knowledge database stores data on the man-hours required for each of the multiple inspection items related to the equipment. The priority setting unit ensures that among the unchecked inspection items, those requiring less man-hours have a higher priority. A maintenance support device according to claim 1, characterized by the following:

5. The diagnostic knowledge database stores data on the failure frequency or failure probability of each of the multiple inspection items related to the equipment. The priority setting unit ensures that among the unchecked inspection items, those with a higher failure frequency or probability of failure receive a higher priority. A maintenance support device according to claim 1, characterized by the following:

6. The diagnostic knowledge database stores data indicating the difficulty level of each of the multiple inspection items related to the equipment. The priority setting unit ensures that among the unchecked inspection items, those with lower inspection difficulty receive a higher priority. A maintenance support device according to claim 1, characterized by the following:

7. The diagnostic knowledge database stores system identification information, including inspection locations, for each of the multiple inspection items related to the equipment. The priority setting unit, when an inspection point for an inspection item that has been most recently checked by a maintenance worker in the checklist is included in a predetermined system, sets the priority of other inspection points included in that predetermined system higher than that of inspection points in systems other than the predetermined system. A maintenance support device according to claim 1, characterized by the following:

8. The priority setting unit, based on the diagnostic history database in which the diagnostic history of the equipment is accumulated, sets a lower priority for inspection items that have not been inspected for a predetermined period of time since the last inspection, compared to other inspection items for which the predetermined period of time has elapsed. A maintenance support device according to claim 1, characterized by the following:

9. The priority setting unit, based on the diagnostic history database in which the diagnostic history of the equipment is accumulated, sets a higher priority than other inspection items for the inspection items that were previously inspected by a skilled maintenance worker and whose inspection order was earliest among the uninspected inspection items. A maintenance support device according to claim 1, characterized by the following:

10. The system includes an inspection progress explanation generation unit that generates explanatory text regarding the progress of the equipment inspection by maintenance personnel, based on the check status of the checklist and the diagnostic knowledge database. A maintenance support device according to claim 1, characterized by the following:

11. The priority explanation generation unit generates the recommended action and the rationale using artificial intelligence (AI). A maintenance support device according to claim 1, characterized by the following:

12. This includes a priority setting process that sets the priority for maintenance personnel to perform unchecked inspection items included in the inspection checklist of the equipment, based on a diagnostic knowledge database in which data on fault diagnosis of the equipment to be maintained is accumulated. A maintenance support method including a priority explanation generation process that generates an explanatory text containing recommended actions that maintenance personnel should take after the most recent inspection, and the rationale for prioritizing those recommended actions over others, based on the diagnostic knowledge database and the priority.

Images